Hi guys,
I'm new to Stata and to this forum. I've got tennis data from ATP matches and I want to run fixed effect regressions. The two basic regressions I want to run are: 1) https://gyazo.com/994caa3fd252bc7cd084df5e47612c54 2) https://gyazo.com/c8d5d321fe6e98283a851975165e4d79
I want to check what can affect the effort that tennis player exert. Effort, which is measured as P, is the number of games a tennis player won in a specific match. I assume that the more games won in a specific match, the more effort the player exerted. HET is the heterogeneity between the two tennis players, measured by the difference in their ATP ranking. Spread is the prize spread for the current round between the winning prize and the losing prize. X should be a set of tournament dummies. My dataset contains, for every single match from 2013, tournament characteristics like surface, which type of torunament (Grand Slam, Masters, ATP 250, ATP 500), etc. I want to control with tournament fixed effect and year fixed effect. For the second regression, there is another variable called "numoftourneys" in which I want to measure the intrinsic motivation of every player. This variable means how many tournaments a specific tennis player participated in a specific year. Now, of course some top tennis players will not participate in the minoric ATP 250 tounaments, and respectively weak players won't play Grand Slams because these tournaments are way out of their league, so in order to control on this difference, I thought about running also a player fixed effect.
However, I truly don't know how to specify those models in Stata. I read about fixed effect modelling in Stata but it made me even more confused. All I did so far is converting string variables to numeric variables, using the egen (group) command. Any help will be much appreciated.
I'm new to Stata and to this forum. I've got tennis data from ATP matches and I want to run fixed effect regressions. The two basic regressions I want to run are: 1) https://gyazo.com/994caa3fd252bc7cd084df5e47612c54 2) https://gyazo.com/c8d5d321fe6e98283a851975165e4d79
I want to check what can affect the effort that tennis player exert. Effort, which is measured as P, is the number of games a tennis player won in a specific match. I assume that the more games won in a specific match, the more effort the player exerted. HET is the heterogeneity between the two tennis players, measured by the difference in their ATP ranking. Spread is the prize spread for the current round between the winning prize and the losing prize. X should be a set of tournament dummies. My dataset contains, for every single match from 2013, tournament characteristics like surface, which type of torunament (Grand Slam, Masters, ATP 250, ATP 500), etc. I want to control with tournament fixed effect and year fixed effect. For the second regression, there is another variable called "numoftourneys" in which I want to measure the intrinsic motivation of every player. This variable means how many tournaments a specific tennis player participated in a specific year. Now, of course some top tennis players will not participate in the minoric ATP 250 tounaments, and respectively weak players won't play Grand Slams because these tournaments are way out of their league, so in order to control on this difference, I thought about running also a player fixed effect.
However, I truly don't know how to specify those models in Stata. I read about fixed effect modelling in Stata but it made me even more confused. All I did so far is converting string variables to numeric variables, using the egen (group) command. Any help will be much appreciated.
Comment