Wall Street Analysts predictions
The skills the authors demonstrated here can be learned through taking Data Science with Machine Learning bootcamp with NYC Data Science Academy.
This study tries to evaluate how accurate Wall Street analysts' quarterly earnings projections are. Wall Street analysts try to guess what publicly listed companies will report for their earnings each quarter, and using CNBC's publicly available data, we try to evaluate their accuracy. Study has been based on 1st and 2nd quarter of 2017, and 4th quarter of 2016 earnings reports.
“Many analysts will incorporate top-down factors such as economic growth rates, currencies and other macroeconomic factors that influence corporate growth. They use market research reports to get a sense of underlying growth trends. To understand the dynamics of the individual companies they cover, really good analysts will speak to customers, suppliers and competitors. The companies themselves offer earnings guidance that analysts build into the models.”
As there are more analysts covering the company, there seems to be less earnings surprises.
It seems that analysts' quarterly earnings are quite accurate. The slope for the regression(beta1) comes out to be 0.96 or close to 1, and the intercept(beta0) is 0.00. R-squared is 0.72.
The implication for the accuracy of the analysts' earnings projection is that it will not be easy to take advantage of earnings surprises for profit, even with access to the actual earnings reported.
The result of this study shows that at least for earnings announcements in the most recent quarters it seems that the company's quarterly earnings are quite well reflected in Wall Street the analysts' earnings projections.