www.crosshyou.info

政府統計の総合窓口のデータや、OECDやUCIやのデータを使って、Rの練習をしています。ときどき、読書記録も載せています。

OECD Nutrient balance data analysis 7 - Simple Regression and Multiple Regression using R

Photo by Harry Gillen on Unsplash 

www.crosshyou.info

This post is following of the above post.

In the previous post, I made scaled variables in df4, let's see correlation matrix of those variables.

The most highly correlated variable pair is s_po_to and s_ni_to. The second is s_ni_kg and s_po_kg.

s_ni_to and s_po_kg are not highly correlated, it is 0.103.

So, I will do regression analysis, one is s_ni_kg = beta_0 + beta_1 * s_po_kg + u, the  other is s_ni_kg = beta_0 + beta_1 * s_po_kg + beta_2 * s_ni_to + u.

The fisrt one is simple regression and the second one is multiple regression.

Let's see scatter plot matrix for those variables.

Then, let's use lm() function for regression anaysis.

7.494e-01 means 0.794.

So, simple regression model is

s_ni_kg = 0 + 0.749 * s_po_kg + u

Next, let's add s_ni_to for explanatory variables.

So, multiple regression model formula is

s_ni_kg = 0 + 0.756 * s_po_kg - 0.065 * s_ni_to + u

let's make scatter plot and regression line.

For multiple regression, I use average value of s_ni_to.

The both scatter plot are very similar shape.

That's it. Thank you!

Next post is

www.crosshyou.info

 

To read from the first post,

www.crosshyou.info