Why do many MATLAB examples model stock price returns instead of absolute changes?

2 views (last 30 days)
Many MATLAB examples about fitting ARIMA models use price2ret to transform a stock time series to a return time series and then apply the estimate function to fit an ARIMA model. Why are price returns fit instead of absolute price changes?

Accepted Answer

Rick Rosson
Rick Rosson on 28 Sep 2014
Edited: Rick Rosson on 28 Sep 2014
Suppose you own a portfolio containing two stocks, with a total value of $5,000 each. Suppose further that the price of one stock rose yesterday by $5.00 per share, while the other one rose by $3.00 per share. Which stock contributed more money to your portfolio as a result of yesterday's market changes?
The question is impossible to answer without knowing either the number of shares of each company you hold, or the initial price per share of each stock.
If, on the other hand, I had told you that the first stock rose by 2 percent while the second rose by 4 percent, you would have known immediately that the second stock had performed better and had contributed more of a net increase in value to your portfolio than had the first stock.
Absolute changes in per share stock prices are meaningless. The change in price relative to the initial price -- in other words, the percentage rate of return -- is what actually matters.
  3 Comments
Rick Rosson
Rick Rosson on 29 Sep 2014
Yes, all valid reasons. In fact, if you ensure that the percentage rate of return is always greater than -100 percent, then the price of the stock will never go negative. And this requirement is independent of the current price of the stock.
Rick Rosson
Rick Rosson on 29 Sep 2014
One other thing to keep in mind: when computing the percentage rate of return, don't forget to include any cash dividends that the company pays in the total return. You should include the cash dividends on the so-called "ex-dividend" date, not on the announcement date or the payment date.

Sign in to comment.

More Answers (0)

Categories

Find more on Conditional Mean Models in Help Center and File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!