Creating a loop for linear regression

Question

I have one data.frame with three columns, Name_of_brand,Price and Quantity. I want to calculate coefficient of linear regression with (lm) function.

    Name_of_brand       Price     Quantitity
1.    Brand 1              80         100
2.    Brand 1              85          95
3.    Brand 2              90          80
4.    Brand 2              90         100
5.    Brand 2             100         100
6.    Brand 3             150          80
7.    Brand 4             155          70
8.    Brand 5             165          70
9.    Brand 5             165          60
10.   Brand 6             170          60
11.   Brand 7             180          60
12.   Brand 7             180          60
13.   Brand 7             180          70
14.   Brand 8             170          80
15.   Brand 8             170          60

First I want to convert figures in log, group by Name_of _brand and after that calculate elasiticity for price for each like example below e.g Brand 1, Brand 2 etc.

Brand 1 Table

Name_of_brand       Price    Quantitity

1. Brand 1 80 100 2. Brand 1 85 95

Brand 2 Table

    Name_of_brand       Price    Quantitity
3.    Brand 2              90          80
4.    Brand 2              90         100
5.    Brand 2             100         100

Brand 3 etc...

And the end I want to get final_table with two columns, first column with Name_of_brand and Coeff_elasticity.

Final_table

    Name_of_brand  Coeff_elasticity.
1.    Brand 1            -0,5
2.    Brand 2            -0,6
3.    Brand 3            -0,7
4.    Brand 4            -0,7
5.    Brand 5            -0,5
etc.

Can anyone help me with some code for calculation?

Answer 1

There is no need to explicitly separate your data into several subsets.

model <- plyr::dlply(data, "Name_of_brand", function(df) lm(log(Quantitity) ~ log(Price), data = df))

You may retrieve the coefficients for each level of "Name_of_brand" using coef:

coefficients <- plyr::dlply(model, coef)