100 likes | 258 Views
Earnings Surprises and Signal Analysis . matt mcConnell David Nabwangu Eskil Sylwan Johnson Yeh. Agenda. Background Hypothesis Methodology Data Fitting Explanatory Variables Regression Results Conclusion. Market Reaction to News . In an Ideal World. Reality.
E N D
Earnings Surprises and Signal Analysis matt mcConnell David Nabwangu Eskil Sylwan Johnson Yeh
Agenda • Background • Hypothesis • Methodology • Data Fitting • Explanatory Variables • Regression Results • Conclusion
Market Reaction to News In an Ideal World Reality In a More Realistic World We expect stock price to react to news like this: Delay Overshoot Settling Time AnnouncementDate
Hypothesis • Abnormal returns after earnings surprises follow a curved pattern which can be modeled using the step response of a second-order system • 6 Curve parameters are predictable using information about the company * Results could be applicable to any news item – earnings are measurable
Methodology • Identify earnings surprises (Factset) • 600 events, 100 companies • Retrieve price and other data series (Datastream) • Calculate abnormal returns in ±30 day window • Fit a curve to each event • Least squares method with solver • 6 parameters for each event • Regress 6 parameters on several explanatory variables
Fitting the Data • In some instances the data fit very well • In some instances fit not good Correlation = 91.6% Average Correlation = 80% Correlation = 70%
Explanatory Variables 6 explanatory variables for curve parameters • Quarterly Earnings Surprise % • Positive influence on magnitude, Zeta • 1-Year Price Growth • Negative Impact on Offset, Positive Impact on Magnitude, and Zeta • Quarterly Earnings Surprise $ • Positive Impact on Magnitude, Zeta • Price to Earnings Ratio • Positive Impact on Offset, Negative Impact on Zeta • Beta • Positive Impact on wm, wd, & Magnitude • 10-Day Abnormal Return • Negative Impact on Magnitude
Conclusion • Holds promise: Some predictive power • Paths forward: • Better fitting method • Least squares method more applicable to linear • Improve predictive regressions • More predictor variables • Non-linear predictor variables • Test predictability over time • Larger data set • Create and test trading strategies