Menu
I am using the dplyr and broom combination (per below) and following Fitting several regression models with dplyr to extract the regression coefficients of regressions by group. However - i am also interested in the R2 value of each individual regression (not only for the total model). i have tried to play wiht the augment and glance functions - but have not been able to extract the R2 values. is there an easy way to do this?
Many thanks in advance!
library(dplyr)
library(broom)
df.h = data.frame(
hour = factor(rep(1:24, each = 21)),
price = runif(504, min = -10, max = 125),
wind = runif(504, min = 0, max = 2500),
temp = runif(504, min = - 10, max = 25)
)
dfHour = df.h %>% group_by(hour) %>%
do(fitHour = lm(price ~ wind + temp, data = .))
# get the coefficients by group in a tidy data_frame
dfHourCoef = tidy(dfHour, fitHour)
dfHourCoef
hour term estimate std.error statistic p.value
1 1 (Intercept) 92.173945687 21.82132710 4.2240302 5.102470e-04
2 1 wind -0.020840948 0.01335945 -1.5600151 1.361653e-01
3 1 temp -0.162495052 0.90573269 -0.1794073 8.596220e-01
4 2 (Intercept) 53.569821889 20.90439474 2.5626105 1.957638e-02
5 2 wind 0.006492773 0.01273038 0.5100220 6.162329e-01
6 2 temp -0.493028932 0.78353239 -0.6292387 5.370978e-01
7 3 (Intercept) 93.949047453 14.55042590 6.4567902 4.483106e-06
8 3 wind -0.010084298 0.01179878 -0.8546902 4.039553e-01
9 3 temp -0.096177966 0.68416185 -0.1405778 8.897647e-01
10 4 (Intercept) 68.429142611 20.37382251 3.3586796 3.497149e-03
501
broom: let's tidy up a bit. The broom package takes the messy output of built-in functions in R, such as lm , nls , or t. test , and turns them into tidy tibbles. The concept of “tidy data”, as introduced by Hadley Wickham, offers a powerful framework for data manipulation and analysis.
glance() returns a tibble with exactly one row of goodness of fitness measures and related statistics. This is useful to check for model misspecification and to compare many models. augment adds columns to a dataset, containing information such as fitted values, residuals or cluster assignments.
The glance () is used for creating a single row summary known as a glance of a model or any other object. Syntax: **glance (x)** where, x: any model or R object whose summary is to be returned This recipe demonstrates an example of a 'linear regression model' and the glance () function is used in R.
Broom::glance worked for me:
dfHourCoef = glance(dfHour, fitHour)
dfHourCoef
Source: local data frame [24 x 12]
Groups: hour
hour r.squared adj.r.squared sigma statistic p.value df logLik AIC BIC deviance df.residual
1 1 0.08223448 -0.01973947 34.02159 0.80642638 0.4619401 3 -102.2460 212.4921 216.6701 20834.44 18
2 2 0.07546305 -0.02726328 36.19379 0.73460277 0.4935356 3 -103.5458 215.0915 219.2696 23579.83 18
3 3 0.02395679 -0.08449245 37.17711 0.22090326 0.8039358 3 -104.1087 216.2174 220.3955 24878.47 18
4 4 0.04916169 -0.05648701 40.38246 0.46533173 0.6352725 3 -105.8454 219.6909 223.8690 29353.38 18
5 5 0.16704225 0.07449138 34.47921 1.80486969 0.1930220 3 -102.5266 213.0532 217.2313 21398.69 18
6 6 0.13615197 0.04016886 41.64294 1.41849921 0.2678774 3 -106.4909 220.9818 225.1599 31214.42 18
7 7 0.01979010 -0.08912211 39.22426 0.18170693 0.8353563 3 -105.2343 218.4687 222.6467 27693.76 18
8 8 0.00171480 -0.10920578 31.29634 0.01545971 0.9846722 3 -100.4927 208.9853 213.1634 17630.30 18
9 9 0.19935534 0.11039483 36.02611 2.24094183 0.1351943 3 -103.4482 214.8965 219.0746 23361.85 18
10 10 0.16950148 0.07722387 36.99197 1.83686463 0.1879532 3 -104.0038 216.0077 220.1858 24631.31 18
.. ... ... ... ... ... ... .. ... ... ... ... ...
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With
