0,000000

0,000000

1.349

0.185

Financial leverage

-0,000243

0,002852

-0.085

0.932

Industry - IT

0,114800

0,158100

0.726

0.472

Industry - Oil

-0,038110

0,148200

-0.257

0.798

Industry - Retail

-0,062980

0,153000

-0.412

0.683

Industry - Telecom

-0,069720

0,177300

-0.393

0.696

N

70

F-statistics

41.71

Probability > F

0.000 ***

R2 adj.

0.882

VIF - test

NPS (loyalty)	Marketing share	Website	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
4.156075	3.420486	2.204247	1.222053	1.217096	1.925627	2.161615	2.147414	2.884141

Table 11 represents regression model with the analysis results of structural capital indicators. In fact, the dependent variable of the model is the performance of the company presented in Tobin's Q. It is important to mention, research and development is excluded from the regression model, due to it shows high multicollinearity value according to the VIF-test. Moreover, it was made the logarithm of the intangible assets because it was collected from the balance sheet in cash equivalent, there was a big difference between values due to the different size of the companies. The distribution of the logarithm of intangible assets is normal. As the result, regression model shows the positive and significant impact of independent variables on dependent. In particular, the intangible assets show significant and positive impact on Tobin's Q, in the case of the rise of the intangible by 1 percentage point, Tobin's Q Ratio will rise by approximately 0.09 percentage. The number of patents, licenses and trademarks shows the most significant impact on the Tobin's Q Ratio. The rise of the number of patents, licenses and trademarks on 1 percentage will lead to the rise of the Tobin's Q on 0.32 percentage.

Profitability > F is significant in the model. R-squared adjusted is 0.54 that is appropriated for the model and explained the possible cases on significant level. The VIF - test shows that there is no multicollinearity between independent variables.

The formula for the regression model of structural capital:

,

Table 13 Regression Model (Structural capital)

Variable	Beta	Std. Error	t value	P - value
(Intercept)	-2.952000	0.659200	-4.478	0,000059***
Log (Intangible assets)	0.090440	0.032820	2.756	0.008693*
Number of patents, licenses, trademarks	0.003199	0.000544	5.883	0.000 ***
Number of employees	0.001274	0.000001	-1.491	0.143712
Financial leverage	-0.009598	0.005443	-1.763	0.085320
Industry - IT	-1.764455	0.528739	-3.337	0.001127**
Industry - Oil & Gas	-1.423632	0.357129	-3.986	0.000233***
Industry - Retail	0.139144	0.272659	0.510	0.711886
Industry - Telecom	-0.625226	0.309237	-2.022	0.027192*
N	70
F-statistics	8.752
Probability > F	0.000***
R2 adj.	0.5389

VIF-test

Log(ia)	Patents	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
1.353791	4.676965	1.138675	1.307119	5.397669	3.120045	1.711493	2.257898

Table 14 Regression models on the indicators of intellectual capital

Model	I	II	III	IV	V	VI	VII	VIII
TWC	0.045*** (0.011)	0.037** (0.012)	0.031** (0.012)	0.027** (0.009)	0.019** (0.005)	0.019** (0.005)	0.017** (0.005)	0.016** (0.005)
Productivity		0.017 (0.011)	0.017 (0.010)	0.008 (0.008)	0.005 (0.004)	0.005 (0.004)	0.001 (0.004)	0.001 (0.004)
Trainings			0.017* (0.006)	0.013* (0.005)	0.009** (0.003)	0.008** (0.003)	0.008** (0.002)	0.008** (0.002)
Web				0.022*** (0.114)	0.006 (0.075)	0.011 (0.079)	0.012 (0.067)	0.015 (0.072)
Loyalty					0.052*** (0.005)	0.051*** (0.006)	0.050*** (0.005)	0.050*** (0.006)
TMC						0.001 (0.006)	0.006 (0.005)	0.007 (0.006)
Patents							0.001*** (0.000)	0.001** (0.000)
Log(ia)								0.001 (0.016)
Leverage	-0.011 (0.007)	-0,011 (0.006)	-0.010 (0.006)	-0.006 (0.005)	-0.005 (0.003)	-0.005 (0.003)	-0.006 (0.002)	-0.005 (0.002)
Employees	0.027 (0.008)	0.012 (0.008)	0.011 (0.006)	0.011 (0.004)	0.015 (0.005)	0.009 (0.003)	0.010 (0.004)	0.007 (0.003)
Ind. IT	0.243 (0.379)	0.375 (0.381)	0.738 (0.380)	0.606 (0.311)	0.338* (0.159)	0.326 (0.169)	0.476 (0.247)	0.500 (0.300)
Ind. Oil	0.033 (0.308)	-0.257 (0.351)	-0.359 (0.329)	-0.212 (0.271)	-0.120 (0.137)	-0.133 (0.151)	-0.460 (0.152)	-0.404 (0.274)
Ind. Retail	0.213 (0.316)	0.279 (0.313)	0.936* (0.380)	0.923** (0.310)	0.328 (0.167)	0.311 (0.184)	0.390 (0.157)	0.415* (0.186)
Ind.Telecom	-0.235 (0.358)	-0,060 (0.368)	0.255 (0.362)	0.183 (0.296)	0.160 (0.150)	0.134 (0.189)	0.014 (0.164)	-0.031 (0.218)
N	70	70	70	70	70	70	70	70
F-statistics	4.199	4.144	5.051	8.929	42.13	37.66	49.92	40.95
Probability > F	0.001**	0.001**	0.000***	0.000***	0.000***	0.000***	0.000***	0.000***
R2 adj.	0.3137	0.3392	0.4266	0.6181	0.9023	0.8998	0.9285	0.9244

There are eight regression models, which were constructed in order to find the most significant and appropriate one. It is important to mention that the indicator R&D was excluded due to high level of correlation with other independent variables. The indicators are included one by one in order to control the impact on the organizational performance.

In the first model the impact of TWC on Tobin's Q Ratio is analyzed. The result shows that there is statistically significant impact of TWC on Tobin's Q, particularly while TWC is increasing by 1 percentage, Tobin's Q is increasing by 4.5 percentage. R-squared adjusted is equal to 31.37 %, which seems to be appropriate indicator for this model.

In the II model the productivity is included. The significance of TWC decreased to 0.037. According to the model the indicator productivity has not significant impact on the organizational performance.

In the III model trainings are included. TWC is the most significant indicator in comparison with productivity and training. Trainings, in turn, has impact on Tobin's. In the case total share of trained employees grows by 1%, the Tobin's Q Ratio increased by 1.7%.

In the IV model the website is included. In fact, the significance of website quality is 0.022. Moreover, the Beta-coefficients of significant indicators - TWC and trainings decreased to 0.027 and 0.013 accordingly, however, the significance remains the same.

In the V model the loyalty variable, which is measured by NPS indicator, is added. As the result, the significance of impact of website variable has decreased and Beta-coefficient has fallen from 0.022 to 0.006. However, the influence of loyalty variable becomes statistically significant and measured by 0.052. Two other indicators, TWC and trainings, which are significant in IV model, stay significant. In addition, it can be seen from the table that the significance of trainings impact is increased.

In the VI model TMC variable is involved. However, it is revealed that it does not have statistically significant influence on Tobin's Q. Nevertheless, three variables from V model remain positively significant and their Beta-coefficients stay almost unchanged.

The VII model present the patent variable addition to the previous model. It is crucial to highlight that the patents, licenses and trademarks influence on Tobin's Q significantly and positively. Moreover, in this model TWC, trainings and loyalty variables remain significant as in the previous model. Therefore, VII model includes four statistically significant and positive variables, which affect Tobin's Q.

In the final model VIII, all eight independent variables are included. Final step was to add logarithm of intangible assets. It does not have the significant impact on the organizational performance. As the result, it can be seen that four significant variables - TWC, trainings, loyalty, patents, which has an influence on Tobin's Q Ratio. This model is chosen as optimal one as it includes all independent variables and shows statistically significant and positive impact of four variables. R-squared adjusted is growing with the number of independent variables of the models. Finally, R-squared adjusted in the VIII model is equal to 92.44%.

All models are tested on multicollinearity and residuals normal distribution. All figures and tables with these tests are provided in Appendix 2.



Conclusion

This research paper is dedicated to the intellectual capital, especially, its influence on the organizational performance, presented by Tobin's Q Ratio. The aim of the research is to construct optimal model that will identify the indicators of intellectual capital, which influence the organizational performance. The research question is what is the impact of intellectual capital (components?) on organizational performance?

During the research, the following tasks were solved. A review of the theoretical background was conducted, based on previous studies of intellectual capital components and evaluation of organizational performance of the company. On the basis of literature review the best method of the evaluation the market capitalizations in the view of intellectual capital was chosen - Tobin's Q Ratio. Then, the data related to the intellectual capital and organizational performance indicators from 70 companies was collected. After that, the Pearson correlation was constructed. To measure the components separately three regression models based on three intellectual capital components were created. Finally, eight regression models based on all indicators of intellectual capital were created. The optimal model of intellectual capital impact on the Tobin's Q Ratio was found. It is important to mention, all models were checked on the VIF-test, the results show that there was no multicollinearity between variables. Moreover, all models are significant with appropriate R2-adjusted and normal distribution of the models is shown. (Appendices)

This paper focuses on the analysis of intellectual capital and its impact on the company's performance. Firstly, due to the lack of a single unique method for defining the company's intellectual capital, a review of intellectual capital components was conducted, and the indicators were chosen of the wide variety of currently available measurements for analyzing the company's intellectual capital. Secondly, this study evaluates the impact of intellectual capital indicators on Tobin's Q Ratio on the example of the developing market - the Russian market. The research interest to the developing markets is currently raise, hence, this study is the first in this area focusing specifically on the components of intellectual capitals separately on the Russian market.

Five hypotheses are checked in order to answer this research question.

H1: There is statistically significant positive relationship between intellectual capital components and Tobin's Q

To check this hypothesis the Pearson correlation was constructed. The results show that there is strong relationship between the share of total wage costs, Net promoter score, total share of marketing costs, website quality, number of patents, licenses and trademarks, research and development and Tobin's Q. Hence, it is mean that the majority of intellectual capital indicators have relationship with Tobin's. The hypothesis is confirmed.

To provide a deep analysis of each intellectual capital components, regression models are constructed for each component separately in order to define the influence. Therefore, the first hypothesis:

H2: There is a positive and significant impact of human capital indicators on the organizational performance.

According to the results, this hypothesis is confirmed. All three indicators, which are chosen to measure the human capital' influence, show positive and significant impact on organizational performance. These indicators are total share of wage costs, labour productivity, total share of trained employees. It is necessary to highlight that the share of costs, which are related to employees, can play a significant role in organizational performance. Basically, investments in employees will lead to increasing motivation, to growth of quality of products or services. Furthermore, labour productivity, which indicates the level of profit per one worker, should be developed inside the company as it represents the human capital performance. Hence, it is important to company's management to develop human resources of the organization and as the result to increase the productivity of employees. One more indicator of human capital, which should be mentioned, is the share of trained employees. This indicator positively affects the Tobin's Q Ratio, therefore can increase organizational performance, particularly market capitalization. For the company's management it is crucial to provide employees with training programs, qualification improvement and skills development. It will allow to organize a team of experienced and talented employees. In addition, training systems lead to decrease of the personnel turnover rate.

H3: There is a positive and significant impact of relational capital indicators on the organizational performance.

The analysis of the impact of relational capital components on the performance of the company shows that Net promoter score has the most statistically significant and positive, while total share of marketing costs and website quality are not significant. That means that customer loyalty presented in NPS has an important role. The relationships with customers are the key in the development of the company, customers create value and are the focus the company, due to the fact that the activities of the company oriented directly on the customers. Hence, to be competitive on the market company should i create different loyalty programs that will increase the Net promoter score. Taking everything into account, the hypothesis is confirmed due to the crucial role of NPS on relational capital, that influence significantly on the Tobin's Q Ratio.

H4: There is a positive and significant impact of structural capital indicators on the organizational performance.

After the analysis of the results it can be said the intangible assets and the number of patents, licenses and trademarks have significant influence on the organizational performance, particularly on the market capitalization. R&D was excluded from the model, due to the high correlation with indicators, according to the VIF-test. This means, that company should invest in intangible assets, as they have the long-term perspective in comparison with tangible assets. The investments in intangible assets may have continuous development that will positively influence on market capitalization and Tobin's Q Ratio. The number of patents, licenses anв trademarks are also show the significant influence. Hence, intellectual property should be developed in the company, to have good financial results. The more company invests in technological development and innovations, the more patents it has. The number shows how much developments do the company has. In addition, the hypothesis about the impact of relational capital is confirmed.

H5: Intellectual capital indicators have a positive and significant impact on Tobin's Q.

The hypothesis is confirmed, indicators, which have statistically significant and positive impact on Tobin's Q, are defined based on the optimal regression model. This model includes all indicators from previous models, which are related to impact of human, relational, structural capital on the organizational performance. Overall, four intellectual capital variables, which considered to be positively and significant are emphasized: total share of wage costs, total share of trained employees, loyalty and patents. Based on the obtained results, these indicators of intellectual capital play the most important role in the increase of organizational performance. Finally, the managing and development of these components of intellectual capital will allow companies to gain success on the market, to enhance the competitive advantage.

A significant methodological improvement of the model in comparison with previous studies is the use of unique indicators as a "proxy" for intellectual capital components, which are not displayed in financial and other reports, but are collected and evaluated independently. Also, in order for the results to be objective, financial expenses and the share of employees were calculated as a percentage. This method, based on analysing the unique indicators, provides a deeper understanding of the actual state of the company's intellectual capital.

There are two main conclusions which were made based on the study results. First of all, intellectual capital has a positive and statistically significant impact on organizational performance of Russian companies. This conclusion corresponds to results of previous researches, which have investigated the role of intellectual capital (Carson, Ranzijn, Winefield, & Marsden, 2004). The development of intellectual capital inside the company will result in effective corporate governance, increasing productivity of employee, innovations implementation, defining long-term perspectives.

Secondly, the most important indicators of intellectual capital in terms of the contribution to the performance of organization are total share of wage costs, total share of employees who passed training programs, the customer loyalty level which is measured by Net Promoter Score, patents, licenses and trademarks. These indicators are considered to be the drivers of the organizational performance growth.

Limitations

Data availability is the key limitation of the research. It is important to mention, that the method of data collection is secondary data analysis. The annual, financial reports, balance sheets, website and presentation for investors were the main sources for data collection. In fact, all types of reports are published in huge enterprises, small and medium enterprises do not publish detailed information, particularly data intellectual capital, in open sources. Moveover, in the study the information about market capitalization was collected, therefore public companies were analyzed, due to trading shares on the market. Hence, the sample size is reduced. Furthemore, mainly big corporation invest money in intellectual capital and its development. Consequently, there was no sense to analyze other types of enterprises.

Furthermore, the research is based on the Russian market, particularly publicly traded companies. It is important to highlight the features of Russian market that can bound the implication of results of this research. First feature is related to characteristics of doing business, Russian legislation and general conditions provided for entrepreneurs. Secondly, cultural differences can affect the development of intellectual capital. For example, Russian corporate culture, mentality, readiness to increase the intellectual capital, the level of openness to innovation will influence the level of the intellectual indicators' growth. Moreover, Russia is related to developing countries, that have other perspectives of intellectual capital growth in comparison with the developed countries. This growth can depend on the level of national economic development, total assets of companies and total revenue level. Finally, it is necessary to emphasize that the price per share is formed based on the market demand. Overall, the demand can be influenced by the company's position, popularity of the brand. It should be mentioned that Russian companies can be not as well known to a wide range of the investors as the foreign ones. Therefore, they can be not so much in the demand and the total market capitalization level can be lower in comparison with foreign popular companies. Nevertheless, it is important to emphasize that the results of the study are the basis for further research and that model can be tested in other markets.

Managerial implications

The research is relevant both from practical and theoretical perspectives. In the beginning, organizations can apply this research model in order form an idea of the company's intellectual capital using the analyzed indicators, and then measure them.

The results will be useful for companies' management in order to determine the role of intellectual capital in the creating value. Companies should focus on the development of the main important indicators of intellectual capital, which were revealed in the research analysis results: the level of wage costs, the share of employees participated in training programs, provided by companies, the customer loyalty level, the number of patents, licenses and trademarks. The growth of these intellectual capital components will lead to increase of organizational performance, presented by Tobin's Q in term of this research, to strengthening of the competitive advantage.

According to the results, the positive significant impact of intellectual capital components on the organizational performance was proved. Therefore, the crucial implication is that companies can create a strategy for increasing the company's value based on the impact of intellectual capital. In addition, all components of intellectual capital, included human, relational and structural, are valuable for the performance and represent as necessary factors for future success, determining long-term perspectives.

Further research

One of the main possible directions for the future research is to analyze other country with the proposed model of analysis and indicators. Secondly, the analysis of theoretical background showed that there is not a theory about six components of intellectual capital: human capital, structural capital, customer capital, social capital, technological capital, and spiritual capital (Khalique, Abu, Jamal & Adel, 2011). This theory is not researched so deeply by scholars; hence, it could be analyzed by researchers in future. Thirdly, the sphere of intellectual capital is developing very quickly. New indicators are appearing with high frequency. Hence, other independent variables could be taken as indicators of intellectual capital. Moreover, the Tobin's Q was taken as the dependent variable defining the organizational performance. There could be used other indicators for measuring the financial results of the company, for example, EVA or VAIC. It could also be an interesting for future researchers to conduct a mixed-method analysis, with the «quant -> QUAL» design. The first part of the research could be based on our calculations, while for the second part in-depth interviews with the people from medium and small enterprises. It will help researchers to analyze companies of different size.



References

1. Beltramino, N.S., Garcнa-Perez-de-Lema, D. and Valdez-Juбrez, L.E. (2020). The structural capital, the innovation and the performance of the industrial SMES. Journal of Intellectual Capital.

2. Boujelbene, A. & Affes, H. (2013). The Impact of Intellectual Capital Disclosure on Cost of Equity Capital: A case of French firms. Journal of Economics, Finance & Administrative Science 18 (34), 45-53.

3. Bontis, N. (2001). Managing organizational knowledge by diagnosing intellectual capital: framing and advancing the state of the field. International Journal of Technology Management, 432-455.

4. Brooking, A. (1997), Intellectual Capital: Core Asset for the Third Millennium Enterprise. London: Thomson Business Press, 348.

5. Brooking, A. (2009), Dream Ticket: Using Intellectual Capital Management as a Strategic Planning Paradigm. 5th Workshop on Visualising, Measuring and Managing Intangibles and Intellectual Capital, 125-187.

6. Buallay, A. (2017). The relationship between intellectual capital and firm performance. Corporate Governance and Organizational Behavior Review, 1(1), 32-41.

7. Carson, E., Ranzijn, R., Winefield, A. & Marsden, H. (2004). Intellectual capital: Mapping employee and work group attributes. Journal of Intellectual Capital, 5(3), 453-463.

8. Cabello-Medina, C., Loґpez-Cabrales, A. & Valle-Cabrera, R. (2011). Leveraging the innovative performance of human capital through HRM and social capital in Spanish firms. The International Journal of Human Resource Management, 22(4), 807-828.

9. Chaney, P.K., & Philipich K.L. (2002). Shredded Reputation: The Cost of Audit Failure. Journal of accounting research, 254-273. doi: 10.1111/1475-679X.00087

10. Chang, W. S. & Hsieh, J. J. (2011). Intellectual Capital and Value Creation - Is Innovation Capital a Missing Link? International Journal of Business and Management, 6(2), 3-9.

11. Choong, K. K. (2008). Intellectual capital: definitions, categorization and reporting models. Journal of Intellectual Capital, 9(4), 609-638.

12. Combs, J. G., Crook, T. R., & Shook, C. L. (2005). The dimension of organizational performance and its implications for strategic management research. In D. J. Ketchen & D. D. Bergh (Eds.), Research methodology in strategy and management, 259-286. San Diego: Elsevier.

13. Cvrtila, L.D., and Kuna, S.F. (2006). Intellectual Capital and Human Resource Management. Third Interdisciplinary Symposium, Interdisciplinary Management Research III, 373-389.

14. Datta, S. K., & De, T. (2017). Role of Relational Capital and Firm Performance: Analysis of a Cluster of Bell-metal Enterprises in a Rural Region in West Bengal, India. Journal of Entrepreneurship & Organization Management, 6(1), 1-6.

15. Dzenopoljac,V. & Shahnawaz, M. (2009). Assessing the Intellectual and Knowledge Based Assets of Organizations: Case of Global Oil and Gas Operations Firms. Conference: AMCIS 2017, A Tradition of Innovation

16. Di Vincenzo, F., Hemphдlд, J., Magnusson, M. and Mascia, D. (2012). Exploring the role of structural holes in learning: an empirical study of Swedish pharmacies. Journal of Knowledge Management, 16(4), 576-591.

17. Dumay J. & Garanina T. (2013). Intellectual capital research: А critical examination of the third stage. Journal of Intellectual Capital, 14(1), 10-25.

18. Edwinson, L & Malone, M. (1997). Intellectual Capital: realizong your company's true value by finding its hidden brainpower. New York: HarperCollin, 240.

19. El-Bannany, (2008). A study of determinants of intellectual capital performance in banks: The UK case. Journal of Intellectual Capital, 14(1), 10-25.

20. Ericson, S. & Rothberg, H. (2015 A Longitudinal Look at Strategy, Intellectual Capital and Profit Pools. Journal of Intelligence Studies in Business, 5(2), 5-13

21. Garavan, T., Morley, M., Gunnigle, P. & Collins, E. (2001). Human capital accumulation: the role of human resource development. Journal of European Industrial Training, 25(2/3/4), 48-68.

22. Hejazi, R., Ghanbari, M & Alipour, M. (2016). Intellectual, Human and Structural Capital Effects on Firm Performance as Measured by Tobin's Q. Knowledge and Process Management, 23(4), 259-273. doi: 10.1002/kpm.1529

23. Helfert, E. (2004). Techniques Financial Analysis. Irwin Professional Publishing, 8th edition.

24. Hendricks, L. (2002). How important is human capital for development? Evidence from immigrant earnings. American Economic Review, 92(1), 198-219.

25. Hosseini M. & Owlia, M.S. (2016) Designing a model for measuring and analyzing the relational capital using factor analysis: Case study, Ansar bank, Journal of Intellectual Capital, 17(4), 734-757.

26. Hsu L., C. & Wang C., H. (2012). Clarifying the effect of intellectual capital on performance: the mediating role of dynamic capability. British Journal of Management, 23(2), 179-205.

27. Huang C.F. and Hshueh S.L. (2007). A study on the relationship between intellectual capital and business performance in the engineering consulting industry: A path analysis. Journal of Civil Engineering and Management, 13(4), 265-271.



Appendix

The script code from RStudio

library (ggplot2)

library (psych)

library (car)

library( readxl)

#Control variables

fl <- dt$FL

assets <- dt$Assets

liab <- dt$Liabilities

equity <- dt$Equity

ebitda <- dt$EBITDA

mcap <- dt$Mcap

nempl<- dt$N_employees

#Human capital

TWC <- dt$TWC

capacity <- dt$Capacity

train <- dt$Trainings

#Relational capital

loyalty <- dt$Loyalty

mshare <- dt$Mark_share

web <- dt$Website

#Structural capital

patents <- dt$Patents

rd <- dt$RD_perc

ia <- dt$IA

ia_ratio <- ia/assets

d2 <- density(ia)

plot(d2)

d3 <- density(log(ia))

plot(d3)

#Dependent variable

tobin <- dt$Tobin

#Normal dist: Dependent var

qqnorm (tobin)

qqnorm (log(tobin))

d <- density (tobin)

plot(d)

d1 <- density(log(tobin))

plot(d1)

#Dummy variables for industries

ind_oil <- as.factor(dt$Oil)

ind_retail <-as.factor(dt$Retail)

ind_it <-as.factor(dt$IT)

ind_telecom <-as.factor(dt$Telecom)

#Descriptive statistics

descript_dataframe <- data.frame(fl,ebitda, nempl, TWC, capacity, train,

loyalty, mshare, web, patents, rd, ia, tobin)

desc <- data.frame(describe(descript_dataframe))

#Multicollinearity

multicol <- data.frame(log(tobin),fl, nempl, TWC, capacity, train,

loyalty, mshare, web, patents, rd, log(ia))

cor_sp <- data.frame(cor(x= multicol))

# General Regression model

model <- lm(log(tobin)~ fl+nempl+TWC + capacity + train + web

+ loyalty + mshare + patents + log(ia) +

ind_it+ ind_oil + ind_retail + ind_telecom)

summary (model)

qqnorm (model$residuals)

shapiro.test (model$residuals)

vif (model)

# General Regression model 1

model <- lm(log(tobin)~ TWC+ fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

# General Regression model 2

model <- lm(log(tobin)~ TWC+ capacity+fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

# General Regression model 3

model <- lm(log(tobin)~ TWC+ capacity+train+fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

# General Regression model 4

model <- lm(log(tobin)~ TWC+ capacity+train+web+fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

# General Regression model 5

model <- lm(log(tobin)~ TWC+ capacity+train+web+loyalty+fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

# General Regression model 6

model <- lm(log(tobin)~ TWC+ capacity+train+web+loyalty+mshare+fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

# General Regression model 7

model<-lm(log(tobin)~TWC+ capacity+train+web+loyalty+mshare+patents+fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

# General Regression model 8

model<-lm(log(tobin)~ TWC+capacity+train+web+loyalty+mshare+patents+log(ia)+rd+fl+nempl+

ind_it+ ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

#Human

model <- lm(log(tobin) ~ TWC + train + capacity + fl+

ind_it + ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

d6 <- density(model$residuals)

plot(d6)

#Structural

model <- lm(log(tobin) ~ patents + log(ia) +rd+fl+nempl+

ind_it + ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

#Relational

model <- lm(log(tobin) ~ loyalty+ mshare+ web + fl+nempl+

ind_it + ind_oil + ind_retail + ind_telecom)

summary(model)

qqnorm(model$residuals)

shapiro.test(model$residuals)

vif(model)

#Export

write.csv2(desc, file = "desc.csv")

write.csv2(cor, file = "cor_exp.csv")

Normal Q-Q plots and VIF tests for regression models



Figure 1. Normal Q-Q plot for human capital regression model

Table 1. VIF test for human capital regression model

TWC	Trainings	Capacity	FL (financial leverage)	Ind_it	Ind_oil	Ind_retail	Ind_telecom
1.750040	2.240116	1.982932	1.124567	2.138049	2.166565	2.719735	2.358663

Figure 2. Normal Q-Q plot for relational capital regression model



Table 2. VIF test for relational capital regression model

Loyalty	Mshare	Web	FL	Nempl	Ind_it	Ind_oil	Ind_retail	Ind_telecom
4.156075	3.420486	2.204247	1.222053	1.217096	1.925627	2.161615	2.147414	2.884141

Figure 3. Normal Q-Q plot for structural capital regression model

Table 3. VIF test for structural capital regression model

Patents	Log(ia)	FL	Nempl	Ind_it	Ind_oil	Ind_retail	Ind_telecom
9.340999	1.394060	1.145751	1.346153	5.408922	7.797686	1.850504	2.831374



Figure 4. Normal Q-Q plot regression model I

Table 4. VIF test for regression model I

TWC	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
1.453994	1.130161	1.153529	1.905303	1.601183	1.575771	2.021025

Figure 5. Normal Q-Q plot for regression model II



Table 5. VIF test for regression model II

TWC	Productivity	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
1.750040	1.9982932	1.124567	1.167932	2.138049	2.166565	2.719735	2.358663

Figure 6. Normal Q-Q plot for regression model III

Table 6. VIF test for regression model III

TWC	Productivity	Trainings	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
1.861860	2.058205	2.262409	1.141119	1.210133	2.285017	2.192242	2.726686	2.472074



Figure 7. Normal Q-Q plot for regression model IV

Table 7. VIF test for regression model IV

TWC	Productivity	Trainings	Web	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
1.770810	2.111899	2.326115	1.279544	1.152883	1.345006	2.150495	2.193553	2.720171	2.362112

Figure 8. Normal Q-Q plot for regression model V



Table 8. VIF test for regression model V

TWC	Productivity	Trainings	Loyalty	Web	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
2.045128	2.157617	2.349729	2.754502	1.966539	1.227463	1.345006	2.150495	2.193553	2.720171	2.362112

Figure 9. Normal Q-Q plot for regression model VI

Table 9. VIF test for regression model VI

TWC	Productivity	Trainings	Loyalty	Web	Mshare	Patents	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
2.075490	2.335404	2.402888	4.194816	2.133908	4.068513	8.551993	1.233641	2.734212	7.668795	3.737791	3.635365	3.919919



Figure 10. Normal Q-Q plot for regression model VII

Table 10. VIF test for regression model VII

TWC	Productivity	Trainings	Loyalty	Web	Mshare	Patents	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
2.075490	2.335404	2.402888	4.194816	2.133908	4.068513	8.551993	1.233641	2.734212	7.668795	3.737791	3.635365	3.919919



Figure 11. Normal Q-Q plot for regression model VIII

Table 11. VIF test for regression model VIII

TWC	Productivity	Trainings	Loyalty	Web	Mshare	Patents	Leverage	Number of employees	Ind_it	Ind_oil	Ind_retail	Ind_telecom
2.075490	2.335404	2.402888	4.194816	2.133908	4.068513	8.551993	1.233641	2.734212	7.668795	3.737791	3.635365	3.919919

Размещено на Allbest.ru