0.03

0.04

0.03

0.03

0.03

0.03

Good

0.33

0.33

0.33

0.34

0.33

0.38***

Normal

0.53

0.50*

0.53

0.50*

0.53

0.50*

Bad

0.10

0.12*

0.10

0.12

0.10

0.09+

Very bad

0.01

0.01

0.01

0.01

0.01

0.002**

Marital status: Married=1 Not=0

0.59

0.61*

0.59

0.61+

0.59

0.60

Living area: urban=1 rural=0

0.73

0.54***

0.73

0.62***

0.73

0.51***

From the Table 4, representing descriptive statistics of cross section in 2016, we can conclude that 9.2% of the interviewees from the monitored sample were dissatisfied with high prices, 4.2% were dissatisfied with quality of the goods, and 5.7% were dissatisfied with assortment. Probably, the proportion has changed from 2011 because more observations have been collected. There are 56% (2% less than in 2011) of women and 58.7% were married individuals in the sample. 10% of observants have finished 9 grades in school, 59% have secondary education, 27% of individuals have higher education, 3% have no education. Comparing education, we can assume that 1% changes in numbers of secondary and education was caused by growing demand for it in society, because better education gives higher opportunity to get a better job and increase life conditions. Also, it may be caused by bigger sample and higher average age, which had been 46 and became 48.5. 69% of interviewees were living in cities of towns in 2016, and 31% were living in rural areas. 3.7% of interviewees has estimated their health as very good, 34% as good, 50% as normal, 10% as bad and 1.1% as very bad. These numbers are a little bit higher than in 2011.

Table 4

Descriptive statistics for cross section of 2016

Variable	Observations	Mean	Std. Dev.
Dissatisfaction
High Prices: yes=1	107,679	0.091782	-
Quality: yes=1	102,149	0.042614	-
Assortment: yes=1	103,738	0.057279	-
Gender: Woman=1 Man=0	134,852	0.559124	-
Age	112,040	48.55403	17.89481
Education
Basic	112,040	0.099572	-
Secondary	112,040	0.594984	-
Higher	112,040	0.274848	-
No basic	112,040	0.030596	-
Health
Very good	111,885	0.0375475	-
Good	111,885	0.345614	-
Normal	111,885	0.50206	-
Bad	111,885	0.103285	-
Very bad	111,885	0.011494	-
Marital status: Married=1 Not=0	111,058	0.587216	-
Living area: urban=1 rural=0	134,852	0.686352	-

Table 5

T-test results for 2016

2016	Price (y1)	Quality(y2)	Assortment(y3)
	0	1	0	1	0	1
	Mean	Mean	Mean	Mean	Mean	Mean
Gender: Woman=1 Man=0	0.57	0.62***	0.57	0.61***	0.57	0.61***
Age	48.82	47.59***	48.82	45.79***	48.82	44.10***
Education
Basic	0.10	0.11***	0.10	0.10	0.10	0.09
Secondary	0.59	0.63***	0.59	0.59	0.59	0.59
Higher	0.28	0.23***	0.28	0.29*	0.28	0.30**
No basic	0.03	0.03	0.03	0.02***	0.03	0.02***
Health
Very good	0.04	0.03***	0.04	0.04	0.04	0.03**
Good	0.35	0.32***	0.35	0.36	0.35	0.40***
Normal	0.50	0.53***	0.50	0.49+	0.50	0.48***
Bad	0.10	0.11*	0.10	0.10	0.10	0.08***
Very bad	0.01	0.01+	0.01	0.01*	0.01	0.01*
Marital status: Married=1 Not=0	0.59	0.60**	0.59	0.62***	0.59	0.61***
Living area: urban=1 rural=0	0.72	0.46***	0.72	0.55***	0.72	0.44***

Student's t-test have shown the statistically significant differences of means for most of the variables, as it is seen from Table 5, thus we can build a probit model to estimate which characteristic variables were more susceptible to each reason of dissatisfaction and how it has changed after sanctions.

The probit model is the model of binary choose.

(3)

In case of this paper six probit models are needed to be build for each dissatisfaction reasons in 2011 and 2016, and three more with interaction terms between all independent variables and dummy of the year (YEAR2011=1 in 2011 and YEAR2011=0 in 2016). Probit model predict the probability of the dependent valuable to express a positive outcome, rather than to predict its actual value. Predicted values say how likeable individual to be dissatisfied with a certain reason rather than satisfied.

(4)

where: biniary dependent variable;

regressor;

control variables.

The coefficients are calculated by the maximum likelihood estimation. The coefficients illustrates the “reaction” of regressors on the equality of the dependent variable to one, but they cannot be interpreted directly. Coefficients show the direction of probability, but not the probability itself. Control variables are regional dummies to alleviate the effect of differences between regions of Russian Federation, when some regions are much poorer, some located too far from big cities, that can significantly change the reasons of dissatisfaction and consequently the coefficients in models.



Results

1 Time series.

Firstly, the regressions on lags was built to check if there are significant lags. The Table 6 represents the autoregressions of differences of Retail trade volume indexes from 1^st to 12^th lags, the Akaike information criterion (AIC) and the Bayesian information criterion (BIC) illustrates the quality of models. Which model has smaller criteria, that model is better, because criteria shows the approximate amount of lost information by a model. The more lags are included in the model, the better it explains the dependent variable in this case, what is concluded from coefficients of determination (R-squared) values, and less information is lost according to AIC and BIC. However, only several lags' coefficients are significant according to p-values; 3^rd lags' coefficients in regressions containing from one to eleven lags are significant on 90% level, 11^th lag's coefficient is significant on the 95% level and 12^th lag's coefficient is significant on the 99% level. If the values of adjusted R-squared, AIC, BIC and p-values are taken together and analyzed, the regression with twelve lags is the best from the list. The difference between models with 11 and 12 lags is bigger than between others if the upper mentioned criteria are compared.

As the adjusted R-squared notably increased between those models, that 12^th lag is more “powerful” than other. To check this hypothesis, the regression only with 12^th lag was built (R13). It is represented in Table 7 in comparison with regression with 12 lags. The difference in BICs is

significant, 653.8 has fallen to 610.4, and despite the model R13 contains only one regressor adjusted R-squared has not fallen considerably, which leads to conclusion that the role of other lags too insignificant, and they should not be included in the regression with CPI. Moreover, the 12^th lag's coefficient became significant on 99.9% level. The seasonal factor is eliminated with inclusion of 12^th lag into the regression.

												Table 6
			Autoregregressions on lags for Retail trade volume index
	R1	R2	R3	R4	R5	R6	R7	R8	R9	R10	R11	R12
LD.Retail trade volume index	0.110	0.0971	0.0791	0.0762	0.0777	0.0758	0.0741	0.0727	0.0751	0.0647	0.0584	0.0386
	(0.0769)	(0.0773)	(0.0770)	(0.0788)	(0.0792)	(0.0796)	(0.0801)	(0.0808)	(0.0808)	(0.0816)	(0.0807)	(0.0799)
L2D.Retail trade volume index		0.116	0.0994	0.0999	0.110	0.112	0.115	0.117	0.125	0.129	0.116	0.0931
		(0.0773)	(0.0768)	(0.0777)	(0.0794)	(0.0797)	(0.0801)	(0.0808)	(0.0810)	(0.0813)	(0.0805)	(0.0783)
L3D.Retail trade volume index			0.142+	0.142+	0.144+	0.145+	0.145+	0.147+	0.149+	0.156+	0.153+	0.127
			(0.0770)	(0.0777)	(0.0785)	(0.0802)	(0.0806)	(0.0812)	(0.0813)	(0.0819)	(0.0808)	(0.0786)
L4D.Retail trade volume index				-0.00501	-0.00268	-0.00687	-0.0194	-0.0209	-0.0273	-0.0260	-0.0145	0.000981
				(0.0783)	(0.0789)	(0.0797)	(0.0814)	(0.0819)	(0.0820)	(0.0825)	(0.0816)	(0.0792)
L5D.Retail trade volume index					-0.0251	-0.0283	-0.0361	-0.0425	-0.0429	-0.0489	-0.0335	-0.00771
					(0.0787)	(0.0793)	(0.0801)	(0.0820)	(0.0820)	(0.0824)	(0.0814)	(0.0792)
L6D.Retail trade volume index						0.0418	0.0359	0.0349	0.0553	0.0554	0.0558	0.0587
						(0.0791)	(0.0797)	(0.0806)	(0.0820)	(0.0823)	(0.0813)	(0.0789)
L7D.Retail trade volume index							0.0699	0.0693	0.0817	0.0991	0.0912	0.0735
							(0.0795)	(0.0802)	(0.0807)	(0.0824)	(0.0812)	(0.0789)
L8D.Retail trade volume index								0.00617	0.0150	0.0244	0.0314	0.0346
								(0.0801)	(0.0804)	(0.0811)	(0.0816)	(0.0790)
L9D.Retail trade volume index									-0.110	-0.103	-0.0854	-0.0317
									(0.0801)	(0.0806)	(0.0800)	(0.0790)
L10D.Retail trade volume index										-0.0764	-0.0638	-0.0374
										(0.0809)	(0.0799)	(0.0778)
L11D.Retail trade volume index											-0.160*	-0.146+
											(0.0799)	(0.0775)
L12D.Retail trade volume index												-0.229**
												(0.0784)
Constant	-0.0480	-0.0447	-0.0556	-0.0599	-0.0533	-0.0447	-0.0459	-0.0515	-0.0482	-0.0438	-0.0656	-0.0552
	(0.129)	(0.130)	(0.129)	(0.130)	(0.131)	(0.132)	(0.133)	(0.134)	(0.134)	(0.135)	(0.134)	(0.130)
Observations	169	168	167	166	165	164	163	162	161	160	159	158
AIC	656.9	653.7	647.0	646.0	644.6	643.0	641.3	640.1	637.0	634.9	627.1	614.0
BIC	663.1	663.1	659.5	661.6	663.3	664.7	666.0	667.9	667.9	668.7	663.9	653.8
R-squared	0.01	0.03	0.04	0.04	0.05	0.05	0.05	0.05	0.07	0.07	0.10	0.15
Adjusted R-squared	0.01	0.01	0.03	0.02	0.02	0.01	0.01	0.01	0.01	0.01	0.03	0.08
F-stat	2.06	2.15	2.54	1.83	1.53	1.36	1.27	1.11	1.20	1.19	1.48	2.08

Standard errors in parentheses ; + p<0.10, * p<0.05, ** p<0.01, *** p<0.001Table 7

Autoregressions of total Retail trade index with 12 and 12^th lags

	(1)	(2)
	R12	R13
LD.Retail trade volume index	0.0386
	(0.0799)
L2D.Retail trade volume index	0.0931
	(0.0783)
L3D.Retail trade volume index	0.127
	(0.0786)
L4D.Retail trade volume index	0.000981
	(0.0792)
L5D.Retail trade volume index	-0.00771
	(0.0792)
L6D.Retail trade volume index	0.0587
	(0.0789)
L7D.Retail trade volume index	0.0735
	(0.0789)
L8D.Retail trade volume index	0.0346
	(0.0790)
L9D.Retail trade volume index	-0.0317
	(0.0790)
L10D.Retail trade volume index	-0.0374
	(0.0778)
L11D.Retail trade volume index	-0.146+
	(0.0775)
L12D.Retail trade volume index	-0.229**	-0.271***
	(0.0784)	(0.0748)
Constant	-0.0552	-0.0748
	(0.130)	(0.130)
Observations	158	158
AIC	614.0	604.3
BIC	653.8	610.4
R-squared	0.15	0.08
adjusted R-squared	0.08	0.07
F-stat	2.08	13.13

Standard errors in parentheses

+ p<0.10, * p<0.05, ** p<0.01, *** p<0.001

The next step is to check the significanse of different lags of the main regressor, which is differences of CPI. The Bayesian information criterion fluctuating between regressions is not significantly changes, thus it is more reasonable to choose model with less variables. Taking into account that only 1^st lags' coefficients are significant on the 90% level (for regressions till 6 lags) and current meaning is significant on 99% level in R1 model, it is reasonable to choose this CPI meanings for regression. Comparing BICs R1 has 3^rd smallest meaning, 638.1, and its regressors explain the dependent variable on 16%. Other models also have lags with significant coefficients, but BIC is not improving noticeably and most of the coefficients are still not significant.

Summarizing everything upperstated, the 12^th lag, taken from autoregression of total Retail trade volume index, the current period total CPI and couple of lags were combined to the final regression. The Table 9 represents 3 alternatives.

													Table 8
				Regressions on Consumer price indexes' lags
	R0	R1	R2	R3	R4	R5	R6	R7	R8	R9	R10	R11	R12
D.Consumer price index	-0.972***	-0.655**	-0.627*	-0.592*	-0.597*	-0.607*	-0.629*	-0.633*	-0.639*	-0.647*	-0.748**	-0.762**	-0.814**
	(0.175)	(0.243)	(0.254)	(0.255)	(0.256)	(0.257)	(0.260)	(0.257)	(0.259)	(0.264)	(0.265)	(0.281)	(0.282)
LD.Consumer price index		-0.461+	-0.583+	-0.665+	-0.642+	-0.628+	-0.611+	-0.562	-0.556	-0.549	-0.406	-0.429	-0.505
		(0.243)	(0.331)	(0.345)	(0.348)	(0.350)	(0.353)	(0.350)	(0.352)	(0.358)	(0.364)	(0.367)	(0.365)
L2D.Consumer price index			0.143	0.206	0.154	0.162	0.161	0.100	0.107	0.0999	-0.0161	0.0659	0.142
			(0.253)	(0.343)	(0.360)	(0.364)	(0.366)	(0.363)	(0.366)	(0.369)	(0.372)	(0.377)	(0.373)
L3D.Consumer price index				-0.0448	0.0325	-0.0184	-0.0492	-0.0152	-0.0259	-0.0241	0.0239	-0.0558	-0.115
				(0.255)	(0.348)	(0.365)	(0.368)	(0.364)	(0.368)	(0.370)	(0.367)	(0.374)	(0.374)
L4D.Consumer price index					-0.0754	0.0824	0.161	0.225	0.231	0.225	0.265	0.315	0.353
					(0.256)	(0.349)	(0.367)	(0.365)	(0.367)	(0.371)	(0.367)	(0.369)	(0.370)
L5D.Consumer price index						-0.178	-0.313	-0.541	-0.530	-0.521	-0.596	-0.584	-0.603
						(0.257)	(0.352)	(0.364)	(0.368)	(0.370)	(0.368)	(0.367)	(0.365)
L6D.Consumer price index							0.137	0.679+	0.638+	0.634+	0.688+	0.668+	0.663+
							(0.259)	(0.349)	(0.367)	(0.370)	(0.366)	(0.367)	(0.363)
L7D.Consumer price index								-0.588*	-0.494	-0.499	-0.446	-0.446	-0.458
								(0.257)	(0.351)	(0.370)	(0.367)	(0.369)	(0.365)
L8D.Consumer price index									-0.102	-0.0494	-0.275	-0.236	-0.192
									(0.259)	(0.357)	(0.371)	(0.372)	(0.368)
L9D.Consumer price index										-0.0648	0.498	0.402	0.363
										(0.262)	(0.360)	(0.374)	(0.370)
L10D.Consumer price index											-0.604*	-0.484	-0.474
											(0.263)	(0.362)	(0.370)
L11D.Consumer price index												-0.107	0.0859
												(0.278)	(0.361)
L12D.Consumer price index													-0.278
													(0.279)
Constant	-0.0926	-0.103	-0.107	-0.127	-0.135	-0.134	-0.125	-0.137	-0.140	-0.136	-0.143	-0.159	-0.147
	(0.119)	(0.119)	(0.120)	(0.119)	(0.120)	(0.121)	(0.122)	(0.122)	(0.123)	(0.124)	(0.123)	(0.124)	(0.123)
Observations	170	169	168	167	166	165	164	163	162	161	160	159	158
AIC	633.4	628.7	627.3	622.2	621.0	619.6	618.1	611.9	611.1	610.1	603.3	600.3	593.4
BIC	639.6	638.1	639.8	637.8	639.7	641.3	642.9	639.8	642.0	644.0	640.2	640.2	636.3
R-squared	0.15	0.17	0.18	0.19	0.19	0.19	0.19	0.22	0.22	0.22	0.25	0.25	0.26
adjusted R-squared	0.15	0.16	0.16	0.17	0.16	0.16	0.16	0.18	0.17	0.17	0.19	0.19	0.19
F-stat	30.75	17.44	11.75	9.26	7.38	6.18	5.34	5.41	4.75	4.24	4.46	4.03	3.90

Table 9

Variations of final regression for total indexes

	(1)	(2)	(3)
	R1	R2	R3
L12D.Retail trade volume index	-0.157*	-0.138+	-0.141+
	(0.0737)	(0.0740)	(0.0742)
D.Consumer price index	-0.892***	-0.603*	-0.551*
	(0.184)	(0.243)	(0.254)
LD.Consumer price index		-0.437+	-0.596+
		(0.243)	(0.330)
L2D.Consumer price index			0.179
			(0.252)
Constant	-0.101	-0.105	-0.102
	(0.121)	(0.120)	(0.121)
Observations	158	158	158
AIC	583.9	582.6	584.1
BIC	593.1	594.9	599.4
R-squared	0.20	0.22	0.22
adjusted R-squared	0.19	0.20	0.20
F-stat	19.29	14.12	10.68

Standard errors in parentheses

+ p<0.10, * p<0.05, ** p<0.01, *** p<0.001

The R1 regression was chosen to be the final, because both coefficients (of 12^th lag and CPI) are statistically significant on thr 95% level (CPI even 99.9%), while the R2 and R3 have less significant coefficients of 12^th lag of Retail trade volume index (90% level) and CPI (95% level) and its lags (90% level), which make them worse comparing to R1. Moreover, the BIC of R1 is the lowest and adjusted R-squared is 1% less than in R2 and R3, which is provided by extra 2 and 3 variables. The coefficients of (differences) total CPI and 12^th lag have negative sign, since in the model differences were used, when they increase, the original meaning decrease. In this case, when CPI falls, the Retail trade index (dependant variable) grows.

For food and nonfood categories the same manipulations with choosing lags were processed in Stata, and results are presented in Appendix 2-7. For food category only the 12^th lag's coefficient is significant at the 99.9%. The model of autoregression with 12 lags shows the best BIC, AIC and adjusted R-squared, but as only the last coefficient is significant, the autoregression with 12^th lag only was built. As in the case with total indexes, this model showed the best results, thus the 12^th lag was included to the final regression for food goods. Absolutely the same situation was with nonfood Retail trade volume index's autoregression. The only 12^th lag was added to final model of this category. The regressions with regressors food CPI and its lags have statistically significant coefficients (at 99.9% level) only for the current meanings of CPI. Since, the Final model represented in Table 10 includes only two regressors; 12^th lag of dependent variable and CPI. Interpretation of the coefficients preserves the same as for total indexes, but they have become statistically significant on 99.9% level.

Table 10

Final linear regression for food indexes and comparative regression.

	Final	Comparative
L12D.Retail trade volume index (food)	-0.255***	-0.261***
D.Consumer price index (food)	(0.0731)	(0.0735)
	-0.406***	-0.335*
LD.Consumer price index (food)	(0.0956)	(0.131)
		-0.185
L2D.Consumer price index (food)		(0.166)
		0.191
Constant		(0.129)
	-0.0775	-0.0741
	(0.112)	(0.112)
Observations	158	158
AIC	558.8	560.5
BIC	568.0	575.8
R-squared	0.21	0.22
adjusted R-squared	0.20	0.20
F-stat	20.77	10.96

However, for nonfood categories regressions with CPI and lags had different results (Appendix 7). The current meaning of CPI (differences) was significant only in case when it was the only regressor in the model. Adding lags to the model has shown that 1^st,2^nd ,3^rd and 12^th lags have statistically significant coefficients (at 99.9%, 99.9%, 90% and 95% level respectively). As the chosen boarder for this study is 95%, thus 3^rd lag was excluded from final regression. The Table 11 represents the variations of final models for nonfood category. According to BIC and AIC the best model is R4, which includes upper mentioned lags of CPI (without 3^rd) and include 12^th lag of Retail trade volume index. Adding one lag has decreased both criteria quite noticeably, and adjusted R-squared has grown to 38%, which is the highest among the final models of all categories.

Table 11

Variations final regressions for nonfood indexes

	R1	R2	R3	R4
L12D.Retail trade volume index (nonfood)	-0.223**	-0.140*	-0.140*	-0.231**
	(0.0710)	(0.0697)	(0.0695)	(0.0719)
D.Consumer price index (nonfood)	-1.665***	-0.185
	(0.314)	(0.455)
LD.Consumer price index (nonfood)		-2.944***	-3.135***	-3.501***
		(0.652)	(0.450)	(0.446)
L2D.Consumer price index (nonfood)		1.585***	1.658***	1.938***
		(0.462)	(0.425)	(0.417)
L12D.Consumer price index (nonfood)				-1.111***
				(0.314)
Constant	-0.0983	-0.0918	-0.0911	-0.131
	(0.165)	(0.156)	(0.155)	(0.150)
Observations	158	158	158	158
AIC	681.2	665.4	663.6	653.2
BIC	690.4	680.7	675.8	668.5
R-squared	0.26	0.34	0.34	0.39
adjusted R-squared	0.25	0.33	0.33	0.38
F-stat	26.68	20.03	26.79	24.72

After the final regressions for total, food and nonfood categories were chosen, the tests on structural break were conducted, and the Hypothesis 1 were tested. The traditional test on structural break is Chow test, but the exact date should be known to divide the time series beforehand. We do not have this date, thus the estat sbsingle test were processed in Stata, this is supremum Wald test for unknown break date. Test is robust to heteroscedasticity. The results for total indexes' regression indicate structural break in January 2015, rejecting null hypothesis (no break) at the 0.026% level. However, the tests for food and nonfood categories failed to reject null hypothesis. This can be interpreted by the error of the test, or for total indexes the significant drop was detected. However, the regression of nonfood category without CPI lags was tested, and the null hypothesis was rejected at the 0.0157% level, but this regression has worse (higher) BIC comparing to the final chosen, so we cannot rely on that. To sum up, the significant drop in 2015 have leveled up and the general declining trend of consumption volume has preserved, which is illustrated by the Figures in Methodology section.

2 Cross sections

For each of cross sections of 2011 and 2016, were build three probit models for each reason of dissatisfaction price (y1), quality (y2) and assortment (y3). The purpose is to identify the direction of reaction's probability of different independent variables on equality of dependent variable to zero. Table 12 and 13 show the results.

First (binary) variable is gender, female = 1 and male = 0, thus the coefficients shows that women are more likely to be dissatisfied with price, quality and assortment of products than man. The coefficients in all three models are statistically significant on 99.9% level. The coefficient for variable Age is statistically significant at the 95% level in Assortment model and have negative sign. Since, we found that older citizens were less likely to be dissatisfied with assortment of goods in 2011 than younger. For other two models Age coefficients are insignificant, thus we cannot make any conclusions. The coefficients for variable Secondary education (ed=2) are statistically significant at the 90% level in model Price, at 99% level in model Quality and at 95% level in model Assortment and their signs are positive. This means that people with secondary education are likely to be more dissatisfied with price, quality and assortment of goods. However, we have several variables describing education. Comparing coefficients of Higher education (ed=3) variable and Secondary education we can state that citizens with Higher education are more likely to be dissatisfied. The coefficients of Higher education are statistically significant at higher level comparing to secondary (99.9% level in Quality and Assortment, 95% level in Price model). We can assume that people Higher education earns more, thereby they are more demanding to products they buy. Good, Normal and Bad health coefficients are statistically insignificant, while Bad Health coefficient are significant at the 90% level in Price model and at the 90% level in Quality product and their signs are positive. Since, individuals with bad health are more sensitive to price level than other categories. The marginal effects of this probit models and the rest are in Appendix 8-9.

Table 12

Dissatisfaction reasons

	Price 2011	Quality 2011	Assortment 2011
	(y1=1)	(y2=1)	(y3=1)
Gender (female=1, male=0)	0.141***	0.118***	0.129***
	(0.0299)	(0.0343)	(0.0344)
Age	-0.00124	-0.00437	-0.0119*
	(0.00481)	(0.00570)	(0.00569)
Age squared	-0.0000205	-0.00000227	0.0000214
	(0.0000493)	(0.0000589)	(0.0000594)
Secondary education (ed=2)	0.104+	0.205**	0.130*
	(0.0548)	(0.0670)	(0.0657)
Higher education (ed=3)	0.151*	0.284***	0.256***
	(0.0602)	(0.0725)	(0.0710)
School (ed=1)	0.113	0.115	0.0670
	(0.0953)	(0.127)	(0.123)
Good health (health=2)	-0.112	0.0414	-0.0666
	(0.0802)	(0.105)	(0.0987)
Normal health (health=3)	-0.0380	0.0160	-0.00684
	(0.0827)	(0.108)	(0.101)
Bad health (health=4)	0.155+	0.192	0.109
	(0.0939)	(0.120)	(0.115)
Very bad health (health=5)	-0.0907	0.265	-0.340
	(0.177)	(0.193)	(0.267)
Marital status (married=1, not married=0)	0.0374	0.0329	0.00303
	(0.0313)	(0.0357)	(0.0355)
Living area (urban=1, rural=0)	-0.394***	-0.233***	-0.508***
	(0.0319)	(0.0390)	(0.0372)
Region dummy	yes	yes	yes
Constant	-1.204***	-1.498***	-0.900***
	(0.157)	(0.192)	(0.180)
Observations	18301	16956	17164

In 2016 women were still more likely to be dissatisfied than men. The gender coefficients are statistically significant at the 99.9% level. The coefficient for variable Age is statistically significant at the 95% level in Price model and have positive sign. Consequently, we can conclude that older citizens were more likely to be dissatisfied with prices of goods in 2016 than younger. For other two models Age coefficients are insignificant, thus we cannot make any conclusions. The coefficients for variable High education (ed=3) are statistically significant at the 90% level in model Price with negative sign, at 95% level in model Quality and at 99% level in model Assortment and their signs are positive. This means that people with High education are likely to be more dissatisfied with quality and assortment of goods and less likely to be dissatisfied with price. Normal, bad and very bad health coefficients are statistically significant at 99.9% level in Price model, with positive sign, which means that individuals with diseases are more likely to be dissatisfied with high prices. This is also relatable to the quality of products dissatisfaction: at 99% level across the citizens estimated their health as bad and very bad. The living area is still significant variable as it was in 2011, and according to the negative sign, rural area inhabitants are more likely to be dissatisfied with high prices, quality and assortment of goods in stores located nearby.

Table 13

Dissatisfaction reasons

	Price 2016	Quality 2016	Assortment 2016
	(y1=1)	(y2=1)	(y3=1)
Gender (female=1, male=0)	0.129***	0.102***	0.106***
	(0.0141)	(0.0186)	(0.0170)
Age	0.00750**	0.000535	-0.00200
	(0.00251)	(0.00324)	(0.00298)
Age squared	-0.000130***	-0.0000709*	-0.0000735*
	(0.0000253)	(0.0000329)	(0.0000307)
Secondary education (ed=2)	-0.00244	0.0366	0.0958**
	(0.0262)	(0.0330)	(0.0324)
Higher education (ed=3)	-0.0522+	0.105**	0.191***
	(0.0294)	(0.0361)	(0.0356)
School (ed=1)	-0.022	0.0910	0.144*
	(0.0492)	(0.0710)	(0.0700)
Good health (health=2)	0.0411	-0.0318	-0.0158
	(0.0421)	(0.0460)	(0.0405)
Normal health (health=3)	0.120**	0.00616	-0.0216
	(0.0439)	(0.0482)	(0.0427)
Bad health (health=4)	0.262***	0.164**	0.0109
	(0.0507)	(0.0568)	(0.0535)
Very bad health (health=5)	0.255***	0.278**	0.104
	(0.0757)	(0.0881)	(0.133)
Marital status (married=1, not married=0)	0.0148	0.0435*	0.0158
	(0.0150)	(0.0194)	(0.0177)
Living area (urban=1, rural=0)	-0.613***	-0.429***	-0.664***
	(0.0150)	(0.0200)	(0.0180)
Region dummy	yes	yes	yes
Constant	-1.068***	-1.431***	-1.201***
	(0.0747)	(0.0960)	(0.0933)
Observations	106637	101152	102712

For accurate comparison between two years last 3 probit models with interaction terms between regressors and dummy variable of the year were built (Table 14).

Table 14

Dissatisfaction reasons

	Price	Quality	Assortment
	y1	y2	y3
YEAR 2011 Ч Gender (female=1, male=0)	0.131***	0.109***	0.113***
	(0.0286)	(0.0330)	(0.0326)
YEAR 2011 Ч Age	-0.000582	-0.00341	-0.00807
	(0.00460)	(0.00546)	(0.00536)
YEAR 2011 Ч Age square	-0.0000281	-0.0000128	-0.0000200
	(0.0000471)	(0.0000564)	(0.0000560)
YEAR 2011 Ч Basic education	0.104	0.145	0.0301
	(0.0898)	(0.120)	(0.111)
YEAR 2011 Ч Secondary education	0.195*	0.322**	0.125
	(0.0828)	(0.113)	(0.102)
YEAR 2011 Ч Higher education	0.232**	0.397***	0.235*
	(0.0867)	(0.117)	(0.106)
YEAR 2011 Ч Good health	-0.205**	0.0229	-0.0681
	(0.0759)	(0.101)	(0.0927)
YEAR 2011 Ч Normal health	-0.162*	0.0123	-0.0168
	(0.0777)	(0.104)	(0.0948)
YEAR 2011 Ч Bag health	0.0425	0.199+	0.113
	(0.0883)	(0.115)	(0.108)
YEAR 2011 Ч Very bad health	-0.186	0.261	-0.297
	(0.169)	(0.186)	(0.251)
YEAR 2011 Ч Marital status (married=1, not=0)	0.0437	0.0283	0.00445
	(0.0297)	(0.0341)	(0.0333)
YEAR 2011 Ч Living area (urban=1, rural=0)	-0.437***	-0.238***	-0.513***
	(0.0292)	(0.0354)	(0.0333)
YEAR 2011	0.253+	-0.0329	0.360*
	(0.148)	(0.193)	(0.180)
YEAR 2016 Ч Gender (female=1, male=0)	0.125***	0.103***	0.104***
	(0.0143)	(0.0190)	(0.0174)
YEAR 2016 Ч Age	0.00826**	0.00131	-0.000349
	(0.00256)	(0.00330)	(0.00304)
YEAR 2016 Ч Age squared	-0.000136***	-0.0000794*	-0.0000876**
	(0.0000257)	(0.0000334)	(0.0000312)
YEAR 2016 Ч Basic education	-0.0319	0.0784	0.128+
	(0.0498)	(0.0722)	(0.0717)
YEAR 2016 Ч Secondary education	-0.0394	0.0993	0.214**
	(0.0471)	(0.0697)	(0.0683)
YEAR 2016 Ч Higher education	-0.101*	0.154*	0.302***
	(0.0494)	(0.0722)	(0.0704)
YEAR 2016 Ч Good health	0.0887*	-0.0210	0.000636
	(0.0441)	(0.0470)	(0.0423)
YEAR 2016 Ч Normal health	0.195***	0.0270	0.00493
	(0.0460)	(0.0493)	(0.0445)
YEAR 2016 Ч Bag health	0.332***	0.187**	0.0473
	(0.0524)	(0.0580)	(0.0554)
YEAR 2016 Ч Very bad health	0.319***	0.273**	0.127
	(0.0772)	(0.0895)	(0.134)
YEAR 2016 Ч Marital status (married=1, not=0)	0.0130	0.0492*	0.00922
	(0.0152)	(0.0198)	(0.0181)
YEAR 2016 Ч Living area (urban=1, rural=0)	-0.555***	-0.425***	-0.630***
	(0.0152)	(0.0206)	(0.0188)
Region dummy	yes	yes	yes
Constant	-1.271***	-1.561***	-1.350***
	(0.0859)	(0.115)	(0.111)
Observations	125134	118786	120597

At first glance, some changes are noticeable; some variables has become significant in 2016, when previously they were not and vice versa. For example, in model Price Age was statistically insignificant in 2011, but in 2016 it has turned significant at 95% level and its coefficient shows that older people are more likely to be dissatisfied with high prices, and statistically significant coefficient of Age squared highlights that there is a specific age of highest dissatisfaction. The attitude towards high prices has changed across the population will different health conditions. If in 2011 people with good and normal health were less likely (negative sign of the coefficient) to be dissatisfied at 99% and 95% level, then in 2016 individuals from all categories of health conditions are likely to be dissatisfied with high prices. Moreover, in 2016 people with bad and very bad health have become more concerned about price and quality of the products; they were more likely to be dissatisfied by those reasons. This can be interconnected with the selectivity of products people with diseases used to buy. Imported goods, especially medicines, some foods and equipment for people with diseases were crucially important. After sanctions and ruble's devaluation prices have increased and import products' substitutes were propose, which causes the dissatisfaction of ill individuals. The education level used to play the role in dissatisfaction (come variables are statistically significant and further discussed) about prices and quality for respondents with secondary education and about prices, quality and assortment for respondents with higher education in 2011. We can assume that respondents with higher education level are able to earn more, thus they are more demanding and critical to what they buy. However, in 2016 the sign of the higher education coefficient is negative in Price model, which means that this category of respondents is less likely to be dissatisfied with high prices, but it still be more likely to be dissatisfied with quality (significant at 95% level) and assortment (significant at 99.9% level). This may be explained by the higher income (in general) than in other categories. The last, what is common for all models, females were still to be more likely to be dissatisfied with prices than man but a coefficients are slightly smaller, probably because males have started to be more concerned about prices, and individuals who lives in the cities and towns are still less likely to be dissatisfied with all three reasons (negative sign), which means that rural areas, which are usually poorer, felt changes deeper and stronger.



Conclusion

The aim of this research was to find the changes in consumer behavior after sanctions imposition. Using data of Federal State Statistics Service about the volume of consumption (Retail Trade Volume Index) and inflation (Consumer Price Index) the linear regression was built to test the presence of structural break. Graphs have shown the significant drop of consumption in 2015, but the common trend of decreasing consumption has been lasting since 2009, thus we could not say exactly, if sanctions have influenced the trend. The tests on structural break for all goods has shown the presence of structural break in January 2015, however the tests for food and nonfood categories separately have not. Thus, we can not rely completely on its results, as the different regressions have opposite results, but for nonfood category the regression with the lowest AIC and BIC was different from food and total regressions. If we use for nonfood category the regression the same set of regressors as for food and nonfood (but with higher AIC and BIC), the test shows the presence of structural break at the same month January 2015. Such results can be interpreted in two ways; there was not a structural break, and test results are incorrect, or the test has “caught” the significant drop of consumption, and, as the consumption has partly recovered, it could be “confused”.

Probit models for cross sections were built to study the consumers' dissatisfaction with retail on the base of “Comprehensive monitoring of living conditions of the population” conducted by FSSS. Two years model's comparison has shown that categories of dissatisfied people have changed. Three reasons of dissatisfaction were chosen from questionnaire and studied in this research. Dissatisfaction with price, quality and assortment were dependent variables. According to marginal effects, the area of living plays the most significant role in individuals' dissatisfaction with retail. Respondents from rural areas are more likely to be dissatisfied with all three reasons. In 2016 marginal effects for urban people are equal to -0.082 for price dissatisfaction, -0.032 for quality dissatisfaction and -0.065 for assortment dissatisfaction and statistically significant at 99.9% level, which means that urban people were less likely to be dissatisfied than people from rural area by prices by 8%, by quality by 3%, by assortment by 6%. As most of the goods are bought in big cities and resold with price markup in smaller, if prices rise in cities, they are most likely to be risen when goods are resold. Moreover, citizens from villages have narrower selection of shopping places than individuals in cities, thus they can not really choose the cheaper prices, better assortment or quality, because of absence of alternatives. Also, there is a high probability that individuals in cities earns more, thus they are less sensitive to prices' growth. Talking about the earnings, we can assume that citizens with higher education earn more, thus they are less likely to be dissatisfied with prices after sanctions, but usually people with higher level of income are more demanding, which explains dissatisfaction with quality and assortment sensitivity. According to all pobit models women are more likely to be dissatisfied with retail (by price, quality and assortment), because women buy more often than man, however, the coefficients of probit models show that their dissatisfaction is slightly less (the same with marginal effects), which can mean that men have noticed the significant changes and expressed their discontent in survey of 2016. Moving forward, age has turned to be significant and older people are more likely to be dissatisfied but only on 0.1%. Health of respondents have less role in price dissatisfaction in 2011, when it was indicated that people with good and normal health are less likely to be dissatisfied, while in 2016 citizens of all types of self-assessment of health have become dissatisfied. Taking 2016 separately respondents with normal health are 1.5% more likely to be dissatisfied with prices, and respondents with bad and very bad health are 3.6% and 3.5% more likely to be dissatisfied according to marginal effects of probit models (Appendix 8-9 ).

To sum up, the consumer behavior has transformed in attitude towards retail. The coefficients in probit models, their significance and even signs of some variables have changed, which points on changes. If we summarize everything, we can say that people with better life conditions (which are assumed to be indicated by higher education, urban living area) are more demanding, thus after sanctions and crisis they are more likely to be dissatisfied with quality and assortment of goods. Also, the attitude of people with all health types have changed towards prices, in 2016 every category was dissatisfied. Dissatisfaction with quality is also relatable to people with worse heath, which can be explained by the absence of some import goods, which are mostly used for healing in Russian Federation. Quality of analogues produced in Russia is in question. The assortment of special products for individuals with different illnesses has never been wide, because the demand is not high, and prices of import goods have always been quite high, thus in both years the coefficients are insignificant. The imposition of sanction has influenced on perception of prices, older individuals are more likely to be dissatisfied. Pensioners have less money than middle age or young individuals, and middle-aged families have more expenses, thus older respondents are more likely to be dissatisfied with high prices.

Results of this research can be used for further studies of sanctions' effect on consumer behavior. The limitations or research and its results are framed by the format and volume of information provided in the public domain. The information about income and loans could be very suitable, however, the data is either not public or provided only in yearly format. Also, the used probit models shows only directions and marginal effects of models represents probability of changes but not changes directly. If more numeric data were available, more opportunities for research and building model have become available.

The inclusion of qualitaitve methods of analysis such as in-depth interviews with individuals of narrower region of researching can also make research deeper, cities like Moscow and Saint Petersburg and regions are needed to be studied seperately. The studies of particular region can reveal hidden problems, for example, in some regions, prices can grow much faster, and the discovery of this fact through interviews with people can help the higher regional authorities to somehow resolve this. In most of the cases the state regulations, support and control are the only instruments to solve such problems. Taking into account the current economic situation, the state at least will know the problems, and if some opportunies appear, then it will know what is needed to be improved. Next population census will be conducted in 2020, the gathered data will be a rich resource for studing the current problems. Adding new data can improve the research, as it contains significantly bigger number of observations than “Comprehensive monitoring of living conditions of the population”.



References

Bauer, R.A. (1960) Consumer Behavior as Risk Taking. In: Hancock, R.S., Ed., Dynamic Marketing for a Changing World, Proceedings of the 43rd. Conference of the American Marketing Association, 389-398.

Berdysheva, E., & Romanova, R. (2017). Rethinking prices during an economic crisis: Calculation as a new mode of consumer behaviour in Russia. International Journal of Consumer Studies, 41(4), 397-403.

Brody, R., & Cunningham, S. (1968). Personality Variables and the Consumer Decision Process. Journal of Marketing Research, 5(1), 50-57.

Cunningham, S.M. (1966). Perceived Risk as a Factor in the Diffusion of New Product Information. In Proceedings of National Conference of American Marketing Association, 698-721.

Dalton, B. (2004). Creativity, Habit, and the Social Products of Creative Action: Revising Joas, Incorporating Bourdieu. Sociological Theory, 22(4), 603-622.

Davydenko, V., Romashkina, G., & Nasyrova, R. (2017). Models of Russian Consumer Behaviour: Retail Under Crisis Pressure 1. Ekonomska Misao I Praksa, 26(2), 697-714.

Dichter, E. (1960). The strategy of desire. (1st ed.). Garden City, N.Y.: Doubleday.

Dutt, P., & Padmanabhan, V. (2011). Crisis and consumption smoothing. Marketing Science, 30(3), 491-512.

Evans, F. (1959). Psychological and Objective Factors in The Prediction of Brand Choice Ford Versus Chevrolet. The Journal of Business, 32(4), 340-369.

Haigh, G., & Fiske, D. (1952). Corroboration of personal values as selective factors in perception. Journal of Abnormal Psychology, 47(2 Suppl.), 394-8.

Howard, J. (1963). Marketing: Executive and Buyer Behavior. New York: Columbia University Press.

Kaytaz, M., & Gul, M. C. (2014). Consumer response to economic crisis and lessons for marketers: The Turkish experience. Journal of Business Research, 67(1), 2701-2706.

Kharas, H. (2011). The Challenge of High and Rising Food Prices. The Brown Journal of World Affairs, 18(1), 97-106.

Kollat, D., Engel, J., & Blackwell, R. (1970). "Current Problems in Consumer Behavior Research". Journal of Marketing Research, 7(3), 327-332.

Levy, S. (1966). Social class and consumer behavior. In Joseph W. Newman, ed., On Knowing the Consumer, New York: John Wiley & Sons, Inc. (pp. 146-160).

MacInnis, D. J., & Folkes, V. S. (2009). The disciplinary status of consumer behavior: A sociology of science perspective on key controversies. Journal of Consumer Research, 36(6), 899-914.

May, F. (1965). Buying Behavior: Some Research Findings. The Journal of Business, 38(4), 379-396.

McCormack, D., Adey, P., Anderson, B., & Graham, S. (2015). Governing Inflation: Price and Atmospheres of Emergency. Theory, Culture & Society, 32(2), 131-154.

Nicosia, F. (1966). Consumer decision processes: Marketing and advertising implications (Prentice-Hall international series in management). Englewood Cliffs, N.J.: Prentice-Hall.

North, D. C., Wallis, J. J., & Weingast, B. R. (2011). Violence and social orders: The conceptual framework for the interpretation of the humanity written history. Moscow: Publishing Gaidar Institute.

Pastore, N. (1949). Need as a determinant of perception. The Journal of Psychology: Interdisciplinary and Applied, 28, 457-475.

Perriman, H. E., Ramsaran-Fowdar, R. R., & Baguant, P. (2010). The impact of the global financial crisis on consumer behaviour. The 2010 Annual London Business.

Popikov, A. (2016). Analysis of crisis in Russia in 2008-2009 and 2014-2016: preconditions, similarities and differences. Internetnauka, 2(10).

Radaev, V.V. (2008). Modern economic and sociological concept of the market. Economic sociology, 1, 20-50.

Sherman, H. (2011). Portrait of a Crisis. Journal of Economic Issues, 45(3), 703-716.

Sheth, J. (1967). A Review of Buyer Behavior. Management Science, 13(12), B718-B756.

Smith, C. W. (2011). Coping with contingencies in equity option markets:

The Rationality of pricing. In J. Beckert & P. Aspers (Eds.), The worth of goods valuation and pricing in the economy (pp. 272-296). Oxford,UK: Oxford University Press.

Urunov, M. (2015). Economicheskii crisis v Rossii: prichiny, mehanizmi razvertyvaniya I vozmozhnie posledstviya. [ Economic crisis in Russia: reasons, mechanisms of development and possible consequences.] Journal Novoi economicheskoi associacii, (2), 186-190.

Voinea, L., & Filip, A. (2011). Analyzing the main changes in new consumer buying behavior during economic crisis. International Journal of Economic Practices and Theories, 1(1), 14-29.

Мамедли, М. О., & Синяков, А. А. (2018). Финансы домохозяйств в России: шоки дохода и сглаживание потребления. Вопросы экономики, 5, 69-91.



Appendecies

Appendix 1

*MacInnis, D. J., & Folkes, V. S. (2009). The disciplinary status of consumer behavior: A sociology of science perspective on key controversies. Journal of Consumer Research, 36(6), 899-914

Appendix 2
			Autoregregressions on lags for Retail trade volume index (food)
	R1	R2	R3	R4	R5	R6	R7	R8	R9	R10	R11	R12
LD.Retail trade volume index (food)	0.0450	0.0420	0.0432	0.0394	0.0423	0.0472	0.0415	0.0401	0.0401	0.0370	0.0189	-0.0205
	(0.0770)	(0.0775)	(0.0779)	(0.0786)	(0.0788)	(0.0792)	(0.0802)	(0.0808)	(0.0814)	(0.0811)	(0.0812)	(0.0779)
L2D.Retail trade volume index (food)		0.0983	0.0891	0.0968	0.109	0.116	0.122	0.122	0.123	0.124	0.121	0.0756
		(0.0772)	(0.0776)	(0.0782)	(0.0786)	(0.0788)	(0.0797)	(0.0807)	(0.0814)	(0.0812)	(0.0805)	(0.0767)
L3D.Retail trade volume index (food)			0.00541	0.00522	0.00481	0.00132	0.00299	0.00722	0.00969	0.0169	0.0154	0.0144
			(0.0776)	(0.0781)	(0.0785)	(0.0790)	(0.0797)	(0.0808)	(0.0819)	(0.0818)	(0.0811)	(0.0767)
L4D.Retail trade volume index (food)				-0.0697	-0.0600	-0.0720	-0.0774	-0.0792	-0.0801	-0.0640	-0.0611	-0.0523
				(0.0778)	(0.0781)	(0.0785)	(0.0794)	(0.0802)	(0.0813)	(0.0816)	(0.0811)	(0.0767)
L5D.Retail trade volume index (food)					-0.0228	-0.0283	-0.0322	-0.0364	-0.0380	-0.0460	-0.0318	-0.0138
					(0.0782)	(0.0783)	(0.0793)	(0.0803)	(0.0811)	(0.0816)	(0.0813)	(0.0770)
L6D.Retail trade volume index (food)						0.117	0.112	0.113	0.114	0.103	0.108	0.143+
						(0.0781)	(0.0788)	(0.0798)	(0.0809)	(0.0810)	(0.0810)	(0.0769)
L7D.Retail trade volume index (food)							0.0411	0.0390	0.0414	0.0472	0.0331	0.0100
							(0.0792)	(0.0799)	(0.0810)	(0.0814)	(0.0809)	(0.0771)
L8D.Retail trade volume index (food)								-0.00344	-0.00268	0.0130	0.00615	-0.0102
								(0.0798)	(0.0805)	(0.0809)	(0.0808)	(0.0765)
L9D.Retail trade volume index (food)									-0.0212	-0.0149	0.00457	0.0258
									(0.0806)	(0.0806)	(0.0804)	(0.0766)
L10D.Retail trade volume index (food)										-0.136+	-0.137+	-0.100
										(0.0806)	(0.0801)	(0.0762)
L11D.Retail trade volume index (food)											-0.144+	-0.132+
											(0.0807)	(0.0764)
L12D.Retail trade volume index (food)												-0.329***
												(0.0770)
Constant	-0.0446	-0.0394	-0.0520	-0.0588	-0.0447	-0.0421	-0.0463	-0.0510	-0.0511	-0.0518	-0.0684	-0.0677
	(0.120)	(0.120)	(0.121)	(0.122)	(0.122)	(0.122)	(0.123)	(0.124)	(0.126)	(0.126)	(0.125)	(0.119)
Observations	169	168	167	166	165	164	163	162	161	160	159	158
AIC	631.4	628.9	626.4	624.6	621.6	618.5	617.2	616.3	615.6	611.8	606.2	585.3
BIC	637.6	638.3	638.9	640.2	640.2	640.2	642.0	644.1	646.4	645.6	643.0	625.1
R-squared	0.00	0.01	0.01	0.02	0.02	0.03	0.03	0.03	0.03	0.05	0.08	0.18
adjusted R-squared	-0.00	-0.00	-0.01	-0.01	-0.01	-0.01	-0.01	-0.02	-0.02	-0.01	0.01	0.11
F-stat	0.34	0.99	0.57	0.61	0.56	0.84	0.77	0.67	0.59	0.82	1.10	2.62

Appendix 3

Comparison of autoregressions with lags

	R12	R12only
LD.Retail trade volume index (food)	-0.0205
	(0.0779)
L2D.Retail trade volume index (food)	0.0756
	(0.0767)
L3D.Retail trade volume index (food)	0.0144
	(0.0767)
L4D.Retail trade volume index (food)	-0.0523
	(0.0767)
L5D.Retail trade volume index (food)	-0.0138
	(0.0770)
L6D.Retail trade volume index (food)	0.143+
	(0.0769)
L7D.Retail trade volume index (food)	0.0100
	(0.0771)
L8D.Retail trade volume index (food)	-0.0102
	(0.0765)
L9D.Retail trade volume index (food)	0.0258
	(0.0766)
L10D.Retail trade volume index (food)	-0.100
	(0.0762)
L11D.Retail trade volume index (food)	-0.132+
	(0.0764)
L12D.Retail trade volume index (food)	-0.329***	-0.340***
	(0.0770)	(0.0740)
Constant	-0.0677	-0.0690
	(0.119)	(0.118)
Observations	158	158
AIC	585.3	574.2
BIC	625.1	580.4
R-squared	0.18	0.12
adjusted R-squared	0.11	0.11
F-stat	2.62	21.15

Standard errors in parentheses

+ p<0.10, * p<0.05, ** p<0.01, *** p<0.001

												Appendix 4
			Regressions on Consumer Price indexes' lags
	R0	R1	R2	R3	R4	R5	R6	R7	R8	R9	R10	R11	R12
D.Consumer price index (food)	-.486***	-.461***	-0.423**	-0.430**	-0.427**	-0.436**	-0.435**	-0.428**	-0.431**	-0.429**	-0.432**	-0.451**	-0.496**
	(0.0930)	(0.126)	(0.132)	(0.135)	(0.135)	(0.135)	(0.136)	(0.138)	(0.137)	(0.139)	(0.140)	(0.154)	(0.158)
LD.Consumer price index (food)		-0.0407	-0.159	-0.168	-0.189	-0.184	-0.185	-0.194	-0.222	-0.225	-0.221	-0.210	-0.288
		(0.126)	(0.170)	(0.181)	(0.186)	(0.186)	(0.187)	(0.189)	(0.189)	(0.190)	(0.192)	(0.200)	(0.202)
L2D.Consumer price index (food)			0.138	0.129	0.161	0.186	0.184	0.189	0.230	0.224	0.220	0.230	0.314
			(0.132)	(0.181)	(0.195)	(0.199)	(0.200)	(0.202)	(0.201)	(0.203)	(0.205)	(0.207)	(0.210)
L3D.Consumer price index (food)				0.0198	-0.0497	-0.0890	-0.0761	-0.0741	-0.0971	-0.0881	-0.0972	-0.108	-0.154
				(0.134)	(0.186)	(0.199)	(0.204)	(0.205)	(0.204)	(0.206)	(0.208)	(0.209)	(0.209)
L4D.Consumer price index (food)					0.0787	0.176	0.154	0.138	0.127	0.122	0.134	0.144	0.170
					(0.135)	(0.186)	(0.200)	(0.205)	(0.204)	(0.205)	(0.207)	(0.209)	(0.208)
L5D.Consumer price index (food)						-0.113	-0.0701	-0.0394	0.0382	0.0365	0.0284	0.0316	-0.0102
						(0.135)	(0.187)	(0.202)	(0.204)	(0.205)	(0.207)	(0.208)	(0.207)
L6D.Consumer price index (food)							-0.0463	-0.102	-0.248	-0.231	-0.231	-0.230	-0.193
							(0.136)	(0.189)	(0.201)	(0.206)	(0.207)	(0.208)	(0.207)
L7D.Consumer price index (food)								0.0580	0.323+	0.289	0.312	0.301	0.285
								(0.138)	(0.189)	(0.203)	(0.207)	(0.208)	(0.207)
L8D.Consumer price index (food)									-0.280*	-0.218	-0.260	-0.242	-0.247
									(0.137)	(0.190)	(0.204)	(0.208)	(0.207)
L9D.Consumer price index (food)										-0.0657	0.0153	-0.0262	0.0297
										(0.138)	(0.192)	(0.206)	(0.207)
L10D.Consumer price index (food)											-0.0867	-0.0232	-0.125
											(0.139)	(0.199)	(0.207)
L11D.Consumer price index (food)												-0.0620	0.169
												(0.153)	(0.200)
L12D.Consumer price index (food)													-0.277+
													(0.156)
Constant	-0.0534	-0.0652	-0.0643	-0.0814	-0.0872	-0.0764	-0.0759	-0.0742	-0.0844	-0.0862	-0.0866	-0.0973	-0.0953
	(0.111)	(0.111)	(0.112)	(0.112)	(0.113)	(0.113)	(0.114)	(0.115)	(0.114)	(0.116)	(0.117)	(0.117)	(0.117)
Observations	170	169	168	167	166	165	164	163	162	161	160	159	158
AIC	610.1	607.7	606.0	601.5	599.8	596.3	595.6	594.9	589.9	589.2	588.2	586.5	581.1
BIC	616.4	617.1	618.5	617.0	618.5	618.0	620.4	622.7	620.8	623.1	625.1	626.4	624.0
R-squared	0.14	0.14	0.15	0.16	0.16	0.17	0.17	0.17	0.19	0.19	0.19	0.19	0.21
adjusted R-squared	0.13	0.13	0.13	0.14	0.14	0.14	0.13	0.12	0.14	0.14	0.13	0.13	0.14
F-stat	27.26	13.80	9.53	7.59	6.17	5.29	4.49	3.88	3.94	3.52	3.20	2.93	2.94
					Appendix 5
			Autoregregressions on lags for Retail trade volume index (nonfood)
	R1	R2	R3	R4	R5	R6	R7	R8	R9	R10	R11	R12
LD.Retail trade volume index (nonfood)	-0.00802	-0.00662	-0.0195	-0.0268	-0.0277	-0.0274	-0.0286	-0.0277	-0.0264	-0.0294	-0.0254	-0.0299
	(0.0774)	(0.0774)	(0.0776)	(0.0787)	(0.0793)	(0.0796)	(0.0802)	(0.0808)	(0.0813)	(0.0818)	(0.0818)	(0.0767)
L2D.Retail trade volume index (nonfood)		0.105	0.107	0.103	0.100	0.102	0.101	0.103	0.104	0.105	0.111	0.0933
		(0.0774)	(0.0771)	(0.0780)	(0.0792)	(0.0796)	(0.0800)	(0.0807)	(0.0813)	(0.0817)	(0.0817)	(0.0761)
L3D.Retail trade volume index (nonfood)			0.0968	0.0986	0.0967	0.103	0.101	0.101	0.0967	0.0988	0.0946	0.0832
			(0.0776)	(0.0780)	(0.0789)	(0.0799)	(0.0804)	(0.0809)	(0.0816)	(0.0822)	(0.0820)	(0.0765)
L4D.Retail trade volume index (nonfood)				0.0352	0.0359	0.0359	0.0303	0.0295	0.0299	0.0270	0.0298	0.0305
				(0.0783)	(0.0789)	(0.0796)	(0.0808)	(0.0814)	(0.0818)	(0.0824)	(0.0824)	(0.0767)
L5D.Retail trade volume index (nonfood)					0.0171	0.0148	0.00873	0.00706	0.00976	0.0104	0.0201	0.0444
					(0.0789)	(0.0793)	(0.0800)	(0.0813)	(0.0819)	(0.0823)	(0.0823)	(0.0767)
L6D.Retail trade volume index (nonfood)						0.0170	0.0182	0.0189	0.0258	0.0288	0.0290	0.0279
						(0.0792)	(0.0796)	(0.0805)	(0.0818)	(0.0823)	(0.0821)	(0.0766)
L7D.Retail trade volume index (nonfood)							0.0616	0.0622	0.0648	0.0728	0.0692	0.0752
							(0.0796)	(0.0801)	(0.0810)	(0.0822)	(0.0822)	(0.0764)
L8D.Retail trade volume index (nonfood)								-0.0135	-0.0157	-0.0114	-0.0194	-0.00250
								(0.0803)	(0.0807)	(0.0816)	(0.0823)	(0.0766)
L9D.Retail trade volume index (nonfood)									-0.0123	-0.0147	-0.0123	0.0358
									(0.0807)	(0.0812)	(0.0815)	(0.0766)
L10D.Retail trade volume index (nonfood)										-0.0289	-0.0259	0.0133
										(0.0811)	(0.0810)	(0.0758)
L11D.Retail trade volume index (nonfood)											-0.0575	-0.0704
											(0.0810)	(0.0754)
L12D.Retail trade volume index (nonfood)

Страница:

•  1 
•  2 
•  3 

дипломная работа "The transformation of consumer behavior: the effect of sanctions" скачать

Подобные документы

• Trends and challenges in management relating to organizational behavior

  Discussion of organizational culture. The major theories of personality. Social perception, its elements and common barriers. Individual and organizational influences on ethical behavior. The psychophysiology of the stress response.
  
  контрольная работа [27,7 K], добавлен 19.11.2012
  

• Intercultural business communication case analysis

  Formation of intercultural business communication, behavior management and communication style in multicultural companies in the internationalization and globalization of business. The study of the branch of the Swedish-Chinese company, based in Shanghai.
  
  статья [16,2 K], добавлен 20.03.2013
  

• Business process modeling

  Investigation of the subjective approach in optimization of real business process. Software development of subject-oriented business process management systems, their modeling and perfection. Implementing subject approach, analysis of practical results.
  
  контрольная работа [18,6 K], добавлен 14.02.2016
  

• Evaluating the effectiveness of the transport system in Indonesia

  Evaluation of urban public transport system in Indonesia, the possibility of its effective development. Analysis of influence factors by using the Ishikawa Cause and Effect diagram and also the use of Pareto analysis. Using business process reengineering.
  
  контрольная работа [398,2 K], добавлен 21.04.2014
  

• Исследование функций и целей организации и разработка рекомендаций по её развитию на примере ООО "X5 Retail Group"

  Природа и сущность функций предприятия. Цели организации: понятие, классификация, характеристика. Описание целевого подхода к управлению. Методы построения дерева целей. Общая характеристика организации "X5 Retail group", ее миссия и основные функции.
  
  курсовая работа [39,6 K], добавлен 17.12.2014
  

• Разработка стратегии развития предприятия на примере "X5 retail group"

  Анализ внутренней и внешней среды организации: PEST-анализ и анализ среды непосредственного окружения. Организация управления, ассортимент и сбытовая политика, производственная деятельность компании. Оценка кадрового потенциала, организационной культуры.
  
  курсовая работа [245,4 K], добавлен 21.04.2015
  

• Корпоративная культура как решающий фактор повышения эффективности бизнеса

  Цели, задачи и уровни корпоративной культуры, ее виды и элементы, средства, методы и этапы формирования. Влияние корпоративной культуры на эффективность функционирования организации на примере компании "X5 Retail Group". Предложения по ее улучшению.
  
  курсовая работа [40,4 K], добавлен 03.04.2011
  

• Управление конкурентоспособностью (на примере розничного торгового предприятия X5 Retail Group в городе Ижевске Удмуртской Республики)

  Классификация и функции розничных торговых предприятий. Уровни конкуренции и типы рынка. Организационно-правовая и финансово-экономическая характеристика деятельности организации. Анализ управления конкурентоспособностью розничного торгового предприятия.
  
  дипломная работа [1,1 M], добавлен 19.09.2021
  

• Функции и цели организации (на примере "X5 Retail Group")

  Сущность перехода от оперативного к стратегическому управлению. Понятие функций и целей организации. Стратегическое управление как реализация целевого подхода в управлении. Классификация целей организации и исследование ее функций. Уровни дерева целей.
  
  курсовая работа [327,6 K], добавлен 31.10.2013
  

• Коммуникации в организации

  Теоретические основы коммуникаций в деятельности современного менеджера. Коммуникационный процесс: понятие, основные элементы, этапы. Анализ системы коммуникаций управления фирмы "Trade Master", предложения и рекомендации по ее усовершенствованию.
  
  курсовая работа [125,0 K], добавлен 19.03.2011
  

Другие документы, подобные "The transformation of consumer behavior: the effect of sanctions"

• главная
• рубрики
• по алфавиту
• вернуться в начало страницы
• вернуться к началу текста
• вернуться к подобным работам