I became interested in this topic because of the impact of the war in Ukraine on the global economy, including the inflation. Food shortages are a major problem in this context, and it is important to check whether a large food company can survive during that economic conditions. While high food prices may benefit the company's revenues, financial returns may not be too strong. I decided to examine the stability of dividends for investors in the company based on the previous periods.

Here are the results of my work

https://github.com/PawelTokarski95/Grain-Food-Company---Valuation-in-R-

One assumption for this work was the use of the APT model to value dividends from companies. Unlike the CAPM model, which serves the same purpose, the APT model is more accurate and scientifically proven.

​

Variables used in the model are:

1. PPI (Producer Price Index)

2. Interest_rates

3. Longterm_debt (of this company)

4. Shortterm_debt (of this company)

​

In fact, I created two models that both involved multiple regression, but the first model was not as good as the second - I noticed that there was a problem with certain parameters. Specifically, the variable PPI was not significant, as shown below.

​

MODEL 1:

​

​

​

​

​

​

​

​

​

​

​

​

​

MODEL 2:

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

 

I also tested the model to check for heteroscedasticity of the residuals. I did not worry about autocorrelation, as this is not a forecasting model.

 

 

 

 

 

 

 

Below is a comparison of the model to actual observations. As you can see, periods when the dividend was underestimated were almost as numerous as periods of overestimation according to the model.

​

​

​

​

​

​

​

​

​

​

​

​

​

​

​

 

​

 

In conclusion, the valuation contains a certain neutrality of dividends. The problem arose when determining relative volatility. Due to a high volatility coefficient (76%), I would not consider buying these stocks.