0
I want to compare beta coefficient observed for a variable from two different models. Is there any method to do this? Let say beta coefficient of X1 is 0.6462 and beta coefficient for X1 in a different model is -0.14372 I want to compare whether the difference of X1 in both models is significant or not? Looking at the coefficient value we can say it's different since one is positive and other is negative, but whether its significantly different or not.
I thought of using t test to do the comparison.
Calculate the standard error of the difference between the two beta weights from different models: SE_diff = sqrt(1/(n1-3) + 1/(n2-3)) Calculate the t-statistic:t = (beta1 - beta2) / SE_diff, where beta1 and beta2 are the two beta weights you want to compare. Calculate the degrees of freedom: df = n - k where n is the sample size and k is the number of predictors in the model. Calculate the p-value: You can use a two-tailed t-test and look up the p-value using a t-distribution table with df degrees of freedom. Alternatively, you can use statistical software or an online calculator to calculate the p-value.
But I don't know whether this is a valid way of doing this. Does anyone have any guidance on how to solve this.
Thank you.
