Not an answer to the question as asked but you may want to look at this previous discussion

https://www.researchgate.net/post/Is_there_some_other_method_other_than_hausman_test_to_decide_which_method_either_fixed_or_random_should_be_adopted_in_panel_data_analysis

and

https://www.researchgate.net/post/Where_can_I_find_good_material_on_the_difference_between_mixed_models_and_gee_models

This paper of which I am a coauthor argues for the flexibility of random effects

https://www.researchgate.net/publication/233756428_Explaining_Fixed_Effects_Random_Effects_Modeling_of_Time-Series_Cross-Sectional_and_Panel_Data

there is quite a lot there about endogeneity; we argue that a Hausman test does not distinguish between the appropriateness of fixed and random effects models ( we show how to get exactly the same results from either)  but indicates that the cross-sectional effect is different from the longitudinal and that you can do so much more with the random effects approach :

Article Explaining Fixed Effects: Random Effects Modeling of Time-Se...

It can be used SPSS or MATLAB software. You can try them

 type "test for endogeneity running fixed effects model in Stata?" in Google and click on Images. you will find what you are looking for. 

https://www.google.com/search?q=how+to+do+hausman+test+for+endogeneity+in+stata&client=opera&biw=1242&bih=599&source=lnms&tbm=isch&sa=X&ved=0CAcQ_AUoAmoVChMIg4iC_Y3lyAIVwoYaCh2b4wnQ#tbm=isch&q=test+for+endogeneity+running+fixed+effects+model+in+Stata%3F