The epistemological role of S&P 500 signal's non- stationarity on investors' dynamic sentiment formation : evidence for investors' prospect theory preferences

Financial signals' abrupt spectral changes in temporal scale correspond to the perception bias on risk which is a well-known epistemologically decision-making factor in behavioral finance research. In this paper, we explored the epistemological role of S&P 500 signal's non-stationarity on investors' sentiment formation, by analyzing the time-frequency dependency of investors' sentiment on the S&P 500 through a cross wavelet transform analysis to further explore their wavelet coherence (co-movements). Results indicate that S&P 500 signal's abrupt spectral changes across time do have a statistically significant impact on the investors' sentiment formation. Moreover, a comparative analysis of the wavelet coherences between the Bullish/Bearish market sentiments and the S&P 500's signal, reveals the investors' dynamic (historical) epistemological bias toward risk, which corresponds with the dynamic prospect theory investment preferences.