Slope-dummy and Interaction Coefficients in Stata

Подписаться 15 тыс.

Просмотров 16 тыс.

50% 1

How to generate, estimate and interpret slope-dummy and interaction variables and coefficients in Stata, in a simple OLS framework.
"dealing with dummy variables: • Stata Tutorial: Dealin...
encoding text/string into dummy variables: • Stata Tutorial: Encode...
Link to "Gentle Introduction to Stata"
www.amazon.com/gp/product/159...
Link to the excellent Introduction to Econometrics Textbook by AH Studenmund:
www.amazon.com/gp/product/933...
Link to Jeffrey Wooldridge Introductory Econometrics Textbook:
www.amazon.com/gp/product/813...

Опубликовано:

18 дек 2018

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 15

@ryanzale9186 3 года назад

Great video. I am currently a student at MSU taking a class with Wooldridge and he is a wonderful professor. However, it always helps to get additional help from youtube.

@mikejonaseconometrics1886 3 года назад

Haha - then you are in very good hands! Very glad to hear that the videos are helpful though, and feel free to let me know which additional topics would be useful.

@gglucs1799 3 года назад

I wish I could take classes with Wooldrige because his econometrics book is very good

@mikejonaseconometrics1886 3 года назад

Follow him on Twitter - it’s the next best thing! @jmwooldridge

@takesuretozooneyi4836 5 месяцев назад

Very insightful. I wonder how the explanation will look like when we interact two dummies for e.g. gender and race etc,

@kevinz6097 4 года назад

Very helpful. Thanks for the tutorial!

@thijmenden1104 4 года назад

Thank you👏

@MAli-jw2eq 4 года назад

Great video, thank you. I was wondering whether 1. These interaction effects are the same as moderating effects. 2. How do we plot the effects for the two categories?

@haiyen2270 4 года назад

Hi. Thanks for the video. I have a question. How do you interpret the coefficient of female variable in the interaction model? Is it possible to just include edu and edu*female in the model, without including var female? So i meant the command is reg wage edu exper educ#female, in order to find the marginal effect of educ of different gender. Thank you.

@dilrubasharmin5289 Год назад

Great video, thank you. I have a question: How to calculate the marginal effect of the interaction term? If I use margins, dydx (*), then it only gives me the marginal effect of each variable. suppose I have 15 variables, and I am interacting with female variable with all of them. how to calculate that marginal effect?

@MultiNarutoGamer Год назад

Does the interpretation of the female dummy change after including the interaction term?

@sanyfin 4 года назад

i have a question. your interaction term (fem_educ) was statistically significant. Does that mean the marginal impact of an extra year of education for females has no impact on wages? Also, if you want to capture total affect of female education on income you were adding beta coeffecients of educ and fem_edu. However, what would be the p-value of this beta coefficient of educ+fem_edu?

@mikejonaseconometrics1886 4 года назад

Good question! When the interaction term is significant on its own, that means that marginal effect of education significantly differs between the categories (male vs. female, in this case). You can use the 'lincom' command in stata to test the significance of the sum of coefficients (www.stata.com/manuals13/rlincom.pdf).

@danielawilson1048 2 года назад

Hi I have a question about an interaction effect for a squared continuous variable, what exactly does that mean ? for example in economics, it is seen that GDP is often squared, why is that?

@mikejonaseconometrics1886 2 года назад

Good question, Daniela! Square or quadratic transformations are used to "linearize" the X - Y relationship, and can capture diminishing or increasing returns effects (the effect of X on Y changes as X changes). There is a second order or second derivative to be considered. Here is another video discussing these models: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-9NAM0OGsDqA.html If there is an interaction between one variable (say Z) and the square of another variable (say X), that will be more complicated, as there will be a third derivative (the effect of X on Y changes as both X and Z change).