Bubble Formation and (In)Efficient Markets in Learning-to-Forecast and -Optimise Experiments

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. (0)10 408 Abstract. This experiment compares the price dynamics and bubble formation in an asset market with a price adjustment rule in three treatments where subjects (1) submit a price forecast only, (2) choose quantity to buy/sell and (3) perform both tasks. We find deviation of the market price from the fundamental price in all treatments, but to a larger degree in treatments (2) and (3). Mispricing is therefore a robust finding in markets with positive expectation feedback. Some very large, recurring bubbles arise, where the price is 3 times larger than the fundamental value, which were not seen in former experiments. Terms of use: Documents inJEL Classification: C91, C92, D53, D83, D84

Adaptive expectations versus rational expectations: Evidence from the lab

“…The literature provides numerous contributions in this field, using both negative (Hommes et al (2007), Bao et al (2012)) and positive (Hommes et al (2005), Bottazzi et al (2011), Anufriev and Hommes (2012)) feedback system. Moreover, a more recent strand of literature propose the so-called "Learning to Optimize" (Bao et al (2013), Bao et al (2014)) in which subjects make the best choice in terms of productions (for a comparison of these methods see Duffy (2010)). Evidence from these experiments suggests that, in general, there is strong coordination in the group and that there is convergence to the rational equilibrium only if we consider negative feedback.…”

Learning to believe in simple equilibria in a complex OLG economy - evidence from the lab

“…The lower their forecast error, the higher their payo. We use the quadratic payo function, as in Bao et al (2016) :…”

“…In the second design a so-called Learning-to-Optimize experiment (LtOE), see e.g. Bao et al (2013Bao et al ( , 2016, we drop the assumption of optimal conditional savings decisions, and ask the subjects to directly submit the savings quantity of a young individual. This savings decision may be based on his forecast of the return on savings P t /P e t+1 , but we do not explicitly elicit return forecasts from subjects.…”

“…Previous experiments show that when agents have to directly submit quantity decisions, the experimental economies display more variability and coordination of subjects' strategies is more challenging than when subjects have to make forecasts, see e.g. Bao et al (2013Bao et al ( , 2016. We therefore consider two designs, a learning-to-forecast (LtF) design, where subjects only forecast the price and the optimal savings decision is computed based on their forecasts, and a learning-to-optimize (LtO) design, where subjects make directly savings decisions.…”

A reply to Rosser and Kirman

“…It then becomes natural to measure the performance of subjects in terms of gains, utility or profits and reward them accordingly. Recently some learning-to-forecast and learning-to-optimize experiments have been run by Bao et al (2013Bao et al ( , 2014, where both tasks had to be performed and the rewards are profits and/or utility. This extends the results of chapter 8 and it appears that convergence in negative feedback experiments becomes slower, while bubbles in positive feedback experiments become more severe.…”