# import data
away = read.csv(file = "data/jjj-away.csv")
home = read.csv(file = "data/jjj-home.csv")# helper function
first = function(x) {
x[[1]]
}# quick hack to get "MP" (just the minutes part)
away$MP = as.numeric(sapply(strsplit(away$MP, split = ":"), first))
home$MP = as.numeric(sapply(strsplit(home$MP, split = ":"), first))# add "where" variable
home$WHERE = "home"
away$WHERE = "away"# combine data
data = rbind(home, away)# subset to relevant columns
data = data[, c("STOCK", "MP", "WHERE")]# fit poisson regression
fit = glm(STOCK ~ WHERE + MP, data = data, family = "poisson")# check results
summary(fit)##
## Call:
## glm(formula = STOCK ~ WHERE + MP, family = "poisson", data = data)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -2.76492 -0.61543 0.04499 0.54849 2.76299
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 0.35978 0.62264 0.578 0.563378
## WHEREhome 0.66433 0.17888 3.714 0.000204 ***
## MP 0.02640 0.02274 1.161 0.245697
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for poisson family taken to be 1)
##
## Null deviance: 60.568 on 32 degrees of freedom
## Residual deviance: 45.403 on 30 degrees of freedom
## AIC: 149.23
##
## Number of Fisher Scoring iterations: 5