Question

Buy-side vs. sell-side analysts’ earnings forecasts. Financial analysts who make forecasts of stock prices and recommendations about whether to buy, sell, or hold specific securities can be categorized as either “buy-side” analysts or “sell-side” analysts. A group of Harvard Business School professors compared earnings forecasts of buy-side and sell-side analysts (Financial Analysts Journal, July/August 2008). Data were collected on 3,526 forecasts made by buy-side analysts and 58,562 forecasts made by sell-side analysts, and the relative absolute forecast error was determined for each.

a. Frequency distributions for buy-side and sell-side analyst's forecast errors (with the sell-side distribution superimposed over the buy-side distribution) are shown in the accompanying figure. Based on the figure, the researchers concluded: “that absolute forecast errors for buy-side analysts have a higher mean and variance than those for the sell-side analysts.” Do you agree? Explain.

b. The mean and standard deviation of forecast errors for both buy-side and sell-side analysts are given in the following table. For each type of analyst, provide an interval that will contain approximately 95% of the forecast errors. Compare these intervals. Which type of analyst is more likely to have a relative forecast error of +2.00 or higher?

Short answer

1. No

2. Buy-side

Step-by-step solution

(a) Buy-side and sell-side analyst

No. Although both data sets have a peak at approximately the same value, the centre of the sell-side data is greater because the buy-side distribution is skewed slightly to the left while sell-side data is symmetric. The sell-side data has a wider distribution than the buy-side data.

(b) Buy-side analyst

Buy-side analyst:

Buy-side is likely to have +2, or the whole interval for the sell-side is negative. It is impossible to have a positive error for the sell-side.