Hi everybody,
I'm learning R and statistics and wanted to use a real case scenario for my
study... however a colleague is convinced my use of the linear regression
function is incorrect.
Could you please help clarify if what I'm doing is wrong? And in particular
why is wrong?
I was thinking to use a simple linear regression as a basic predictive
model.
I've a number of stores in a region, for each store I know the Tot Revenue
in a given year & the Tot Active Customers for that year. I want to use the
Tot Active Customers as my independent variable x and Tot Revenue as my
dependent variable y.
So for example:
StoreName | Tot Revenue | Tot Active Customers
Store A | 200,000 | 120
Store B | 230,000 | 129
Store C | 220,000 | 119
The sample data has about 65 stores in total. I don't know the average
transaction value or if a customer has transacted more then once.
> LineBestFit = lm(TotRevenues ~ TotActiveCustomers)
> plot(TotRevenues ~ TotActiveCustomers)
> abline(LineBestFit)
I've plotted the data and the line and I get a strong positive linear
pattern with a couple of outliers , nonetheless the plot shows that the more
active customers I've in a store the more revenues (which is expected).
Now my objective is to calculate the slope b (steepness of the line) so that
I can say that for x active customers I've y increase in revenues, and
consequently attempt to predict targets for new stores...is this right?
Any help would be appreciated,
L
--
View this message in context:
http://r.789695.n4.nabble.com/Can-I-use-a-simple-linear-regression-in-this-situation-tp4553384p4553384.html
Sent from the R help mailing list archive at Nabble.com.
