linearregression-go

README

1. Install Golang, git, setup $GOPATH, and PATH=$PATH:$GOPATH/bin
2. go gets

   go get github.com/gocarina/gocsv
   go get github.com/gonum/plot
   go get github.com/gonum/plot/plotter
   go get github.com/gonum/plot/plotutil
   go get github.com/gonum/plot/vg
   go get github.com/pkg/errors
   go get github.com/sajari/regression
   

3. run the source go run main.go
4. review the results

N = 1551
Variance observed = 5.138608046119863e+06
Variance Predicted = 2.623877776398886e+06
R2 = 0.5106203378131093

Bobs Data (int=2017): 4401
Bobs Data (int=2018) Prediction: 5707

Notes: This script predicts the amount of data that will be created based on fitting a sample data set to a linear regression model.

data-createdate.csv was created from relational database that simply has a date timestamp of when data was created. the creation of data is - in theory - in a linear relationship between the dates that data has been created and time.