Skewness is an important concept when it comes to analyzing data and making predictions about data. Skewness is the degree of asymmetry observed in a probability distribution; specifically, skewness is the measure of the degree to which a distribution deviates from the normal distribution, which is symmetrical in shape. Skewness can be either positive (right-skewed, or skewed to the right) or negative (left-skewed, or skewed to the left).

Positive skewness means that the data is concentrated on the left of the mean and that it tails off more slowly as it reaches to the right. Negative skewness means that the data is concentrated on the right of the mean and that it tails off more slowly as it reaches to the left. A normal distribution (bell curve) exhibits zero skewness, which means that the data is centralized around the mean and that it tails off in both directions evenly.

When analyzing data that is asymmetrically distributed, investors need to be aware of skewness to help inform predictions and forecasts. In particular, right-skewness in a return distribution better represents the extremes of the data set, as it reflects an unevenness between the higher values and the lower values. In other words, if a dataset is right-skewed, it has more extreme high values compared to the number of extreme low values. In contrast, left-skewness indicates that the data distribution has more extreme low values compared to the number of extreme high values.

Skewness is an important concept for investors to consider when assessing the risk of a dataset. However, skewness alone does not tell the user the number of outliers in the data set. By understanding skewness, investors are able to understand the direction of the outliers and make more informed decisions.

Skewness is often found in stock market returns as well as the distribution of average individual income. By understanding how skewness affects data distributions, investors and economists can better understand, analyze and predict the outcomes of their datasets.