The study introduces a new model for stock returns that considers how the data can be skewed. By using a z distribution, the model can accurately capture the skewness and kurtosis often seen in stock return data. The researchers found evidence of skewness in U.S. stock returns after World War II. They also discovered that small positive news has less impact on future stock returns than no news at all. Additionally, a model without skewness tends to overestimate future stock returns and variance.