Now in this video we'll be discussing about the different assumptions arise to regression model and the goodness of fit for logistic regression. So, let's start with assumptions. First, the first assumption for registration is the dependent variable should be dichotomous in nature as you know the dependent variable is binary nature or it is dichotomous in nature that is, it is calculating the probability for for Y equals to one that is how probability for Y equals two even and since it is calculating probability the probability value lies between zero and one therefore, the dependent variable value also lies between zero and zero denotes that it is an impossible event and one only note it's ensuring that the error terms do not need to be normally distributed, there should be minimum multicollinearity for the independent variables that is the independent variables should get less influenced by each other. multicollinearity means when the independent variables are very much influenced by each other when we say it is minimum multicollinearity that means, the independent variables should not be that much influenced by each other.

So, mosquito city is not required for logistic regression model is that this is the next assumption then last integration assumes linearity of independent variables in log odds So, log odds industry regression model gives a linear model we will understand this concept when we discuss about odds and odds ratio in the next video where we will understand that Nova cause gives a linear model this assumption we will discuss again in our next video when we'll be doing when we'll be discussing more often odds ratio plus t regression typically requires an adequate sample size that is the sample size should be large or moderate. Now, let's discuss about the goodness of fit for logistic regression. The goodness of fit for logistic regression not to do test nor to test for the goodness of fit lies to division model we do a live show Goodness of Fit Test where my age not null hypothesis is the model is a good one.

An alternative hypothesis or h1 is the model is not good. So the hospital and so Goodness of Fit Test is a Goodness of Fit Test for logistic regression, especially for risk prediction models. A Goodness of Fit Test tells you how well your data fits the model specifically the HL test or hospital issue test calculates if the observed event rates match the expected event rates in population subgroups. So maybe not his model is a good fit h1 alternative hypothesis model is not good H naught means non hypothesis H minus alternative hypothesis. So if my P value is greater than my level of significance, then I will accept the null hypothesis address otherwise I will reject Okay in this video we will be learning till here now in our next video, we will be discussing about odds and odds ratio. So for now, let's end this video here.

Thank you. Goodbye and see you all for the next video.