README
¶
Profiling the data
see that not all of our variables are normally distributed (that is, they are in a bell shape). The sales might be somewhat bell-shaped, but the others do not look to be normal.
Choosing our independent variable
Visually exploring the correlation between the dependent variable and all of the choices that you have for independent variables.
both Radio and TV appear to be somewhat linearly correlated with Sales. The linear relationship with TV seems most obvious
sales = m*TV + b
Train the model
Predicted = 7.0688 + TV*0.0489
Visualize regression line
Multiple regression
Sales = m1 * TV + m2 * Radio + b
Documentation
¶
Index ¶
Examples ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
func DataSummary ¶
DataSummary prints summary statistics of data 1) Profiling the data
func PlotHistogram ¶
PlotHistogram plots histogram of data 1) Profiling the data
Example ¶
PlotHistogram(data.GetAdvertisingCSVData())
func PlotRegressionLine ¶
func PlotScatter ¶
Example ¶
PlotScatter(data.GetAdvertisingCSVData())
func TrainModel ¶
Example ¶
// The mean sales value was 14.02 and the standard deviation was 5.21. // Thus, our MAE is less than the standard deviations of our sales values // and is about 20% of the mean value, and our model has some predictive power. TrainModel(data.GetAdvertisingCSVData())
Output: Regression Formula: Predicted = 7.0688 + TV*0.0489 MAE = 3.01
func TrainModelMultipleIndependentVariable ¶
Example ¶
// Our new multiple regression model has improved our MAE! // Now we are definitely in pretty good shape to predict Sales based on our advertising spends TrainModelMultipleIndependentVariable(data.GetAdvertisingCSVData())
Output: Regression Formula: Predicted = 2.9318 + TV*0.0473 + Radio*0.1794 MAE = 1.26
Types ¶
This section is empty.
Source Files
Click to show internal directories.
Click to hide internal directories.