Hello,
I am relatively new to this and still have trouble deciding which statistical analysis to use. I am currently working on a project in which I want to see if there is a strong relationship between an F1 team's budget and their finishing position in the championship as well as see if there are other factors that determine their positions and if budget helps these other factors. I have data from 10 teams across 5 years (2015-2019). I have also collected variables like budget, position, points, price per point, number of podiums, number of DNF-s, quality of a driver(numerical variable 1-5), and average speed. Now I want to see if budget is the only and/or most important factor in the teams finishing positions, I also want to see if it's true that the higher the budget the higher the number of podiums, the better the quality of drive, etc. My first thought was to use panel data analysis because I have multiple years and multiple teams I want to examine. But I get lost in this because I find it very hard to understand. To me, it's logical to use the fixed effects model but when I test this STATA tells me to use the random effects model. Also, I saw that normally should be satisfied to perform this model analysis, but I have not been able to accomplish it. I tested it, and it came back as non-normal so I transformed the data using the log option (I transformed every variable, adding a 0.001 to values that were 0), I don't know if I need to transform this data and if it is possible just to transform the variables that don't follow the normality or every variable except the dependant. I am asking for any advice regarding the model I have chosen or if I have chosen the correct one how to perform it. Thank you in advance!
These are my variables:
I am relatively new to this and still have trouble deciding which statistical analysis to use. I am currently working on a project in which I want to see if there is a strong relationship between an F1 team's budget and their finishing position in the championship as well as see if there are other factors that determine their positions and if budget helps these other factors. I have data from 10 teams across 5 years (2015-2019). I have also collected variables like budget, position, points, price per point, number of podiums, number of DNF-s, quality of a driver(numerical variable 1-5), and average speed. Now I want to see if budget is the only and/or most important factor in the teams finishing positions, I also want to see if it's true that the higher the budget the higher the number of podiums, the better the quality of drive, etc. My first thought was to use panel data analysis because I have multiple years and multiple teams I want to examine. But I get lost in this because I find it very hard to understand. To me, it's logical to use the fixed effects model but when I test this STATA tells me to use the random effects model. Also, I saw that normally should be satisfied to perform this model analysis, but I have not been able to accomplish it. I tested it, and it came back as non-normal so I transformed the data using the log option (I transformed every variable, adding a 0.001 to values that were 0), I don't know if I need to transform this data and if it is possible just to transform the variables that don't follow the normality or every variable except the dependant. I am asking for any advice regarding the model I have chosen or if I have chosen the correct one how to perform it. Thank you in advance!
These are my variables:
Attached Files
