 इरे अदी उी स्कॉलिनाताह त efforts of the residual correlation of different models, and if resolute były these balls then it would mean 16 is the best model, on the other hand we have to move on simple OLS approach, second Creating Dignostics of current models is मल्टिकौलिन्यारेटी is another assumption of like your basic regression और उस में हम केतने है के you know that multi-colinearity is basically strong association among the independent variables of your model अब कितने माडल जीुस का रहे हैं उन में अगर strong correlation exist करती है तो it means that there is multi-colinearity और हमने उसको minimise करना होता है क्योंके जब 2 या 3 variables strongly correlate का रहे होते हैं हम उनको merge का सकते हैं हम उनको सकत्र क्योंका सकते हैं हमारी ताके हम आपना जो भीटा की values हैं वो आपकी efficient भी होजें वो आपकी reliable भी होजें और उसली उनको हम best estimator के बना सकते हैं तो इसली मल्टिकौलिन्यारेटी कलिये you have the idea VIF value if it is less than 10 it means there is not the problem of multi-colinearity. independent variables are correlated but if they are strongly correlated then we have to minimize their correlation level in independent variable उसके लिए जे again the research paper on Nigerian economy has discussed about the like multi-colinearity its measurement and the results of multi-colinearity and I am going to share with you again the same paper on Nigerian economy ती स्तुड़ ती सेक्षन फीपवाई ती in which you have the test of multi-colinearity and test of multi-colinearity like there are different approaches you can directly check the multi-colinearity by correlation of your independent variable and if they are strongly correlated it means there is a problem of multi-colinearity. on the other hand you have variance inflation factor or variance inflation factor वी इट्व्ग्ध़ you have जी आप को बतारे एन the threshold level is 10 वो खेर है कर the maximum acceptable VI value is 10, 10 से वृपर जाड़ा है than that is not acceptable and we have to minimize the multi-colinearity इस तेख्बार आश छ़ी से कमसार्ण तेख चोड़ दा रज़्ाद of VIF in table number 4 for all the three models and in table number 4 they have the R-Scale value and you have the idea that R-Scale is basically the explained variation in your dependent variable which is caused by an independent variable. इसेतला यह के ल़ेकित्सित्रे and a mean vi f, in agriculture model vi f is only 4.67, which is less than 10. In industrial model, it's only 5.25. In real estate and  disenton truction । । । । it is only 4.65 । । । and for commercial it is 4.72 । । as far as the services are concerned it is 5.35 । । । । । it means । । । that in your 5 of 5 models । । । the VIF is a multiculinearity । । and it is very less than threshold levels । । । । it means there is not the problem of multiculinearity । । । if the value of VIF 10 is greater than 10 । । । at this time we have to rethink । how to minimize the multicollinearity in your model for efficient estimation of the model.