*Policies for Improving the Efficiency of Innovative Clustering in an Emerging Market DOI: http://dx.doi.org/10.5772/intechopen.112150*

coming from the innovation development strategy, allocated territories for innovation development, the regional law on innovation, the innovation support program, the coordinating body for innovation policy, the regional institute for innovation development, the share of allocations for science in the regional budget, the share of the federal budget in the costs of technological innovation, the share of the regional budget in the cost of technological innovation, the number of innovative projects that received federal support, the number of federal development institutions that support innovative projects, the federal funding of innovative projects, the number of territories for innovative development with federal status, and the number of objects of innovative infrastructure to support SMEs. The results of the analysis are shown in **Table 12**.

### **Multiple linear regression on detailed indicators of Russian clusters**

lm(formula = i\_index innov\_strtg\_dev\_5\_1\_1 + ded\_ars\_innov\_dev\_5\_1\_2 + reg\_innov\_law\_5\_1\_3 + sup\_prog\_innov\_5\_1\_4 + coord\_bdy\_innov\_pol\_5\_2\_1 + reg\_instit\_innov\_dev\_5\_2\_2 + apprp\_shr\_sci\_reg\_ budg\_5\_3\_1 + shr\_fed\_budg\_tech\_innov\_csts\_5\_3\_2 + shr\_reg\_bdg\_rech\_innov\_csts\_5\_3\_3 + num\_innov\_ proj\_recid\_fed\_sup\_5\_4\_1 + num\_fed\_instit\_dev\_sup\_innov\_proj\_5\_4\_2 + fed\_fund\_innov\_proj\_5\_4\_3 + num\_innov\_terris\_dev\_fed\_stats\_5\_4\_4 + num\_obj\_innov\_infra\_sme\_sup\_5\_4\_5, data = clustdata\_1)


### **Table 12.**

*Fifth group's variables' influence on Russia's innovation index.*

Of all the predictors of this group, the potential minimum statistical significance of 0.0790 has the number of federal development institutions that support innovative projects. This variable raises the variable used as the output by 0.277054 when other factors remain unchanged. It is also decided to accept that the predictors are associated with the dependent variable based on the p-value.

When looking at the adjusted R-squared value (0.6624), it can be observed that the fitted model explains 66.24% of the statistical relationships of the grouped variables of the linear regression. However, the model's p-value (0.05324 > 0.05) does not demonstrate confidence in rejecting the null hypothesis about the absence of predictor effects on the final result, which casts doubt on the statistical reliability of the model.
