************************* MRBANKS EXPERIMENT DATASET **************************

The data stored here come from a lab-in-the-field experiment, called Mr. Banks,
where volunteers are given a controlled set of financial information -based
on real data from worldwide financial indices- and they are required to guess
whether the market price would go up or down in each situation. Data is used
to explore basic statistical traits, behavioural biases and emerging
strategies. More details can be found in the following publication:

    Mario Gutiérrez-Roig, Carlota Segura, Jordi Duch and Josep Perelló. Market
    Imitation and Win-Stay Lose-Shift strategies emerge as unintended patterns
    in market direction guesses. PLoS ONE. (2016)

For any question or comment please contact any author of the paper.

You can also find more information of the prohect at www.mr-banks.net


EXPERIMENTAL SETTING

The experiment was carried out inside the context of DAU Festival, a board game
fair held in Barcelona during the weekend of 14th and 15th of December 2013.
The event was organized by the Institute of Culture of the City Council and
attracted 6,000 attendants from Barcelona and its surroundings. The experiment
is framed inside the Pop-Up Experiment concept. Participants did not know in
advance the details of the experiment and were asked to play with Mr. Banks
(for the participants the experiment was referred as a game) via an interface
specifically created and accessible through identical iPads only available in a
controlled area -a space with chairs isolated from the rest of the festival-.
At least three researchers simultaneously supervised the experiment at all
times, preventing any interaction among the volunteers and avoiding that
anybody was repeating the experiment. In order to satisfy privacy issues, all
personal data about the participants were anonymized and de-identified in
agreement with the Spanish Law for Personal Data Protection and the
institutional review board and data protection commissioner of the Universitat
de Barcelona. An online informed consent was given by participants for their
clinical records to be used in this study.


DATASET CONTENT

The MrBanks dataset file contains all recordings needed for reproducing the
experimental part of the paper. 283 volunteers were recruited to participate
in the experiment in the DAU Festival of Barcelona. The participants took
18,436 valid decisions (89 times they ran out of time) and made 44,703 clicks
on the screen.


DATASET STRUCTURE

The dataset is stored in a XML file structured in 4 different tables:

    - Series: This table shows the information about market series.
      The table is composed in the following fields: 
        1) id		ID of the entry.
        2) series	Series ID.
        3) index	Real Series Index.
        4) round	Day number (from the initial date).
        5) date		Corresponding date in the real series.
        6) price	Closing price of the corresponding day.
        7) diff		Difference in percentage.
        8) result	Up (1) or down (-1) with respect previous day price.
        9) expert	Expert advice up (1) down (-1) with 60% of success.

    - Users: This table shows the information about users. The table is
      composed in the following fields:
        1) id			User ID
        2) gender		Tag for males (h) and females (d)
        3) age_range		Age range: 0-15 (re0), 16-25 (re1), 
				26-35 (re2), 36-45 (re3), 46-55 (re4),
				56-65 (re5), beyond 65 years old (re6).
        4) education_level	Education Lelvel: No studies (ne0), Primary
				(ne1), Secondary (ne2), High School (ne3),
				University (ne4) and Unavailable (ne5).
        5) question_1		(Question before playing): Have you
				participated in other citizen science
				experiments? (Yes=0, No=1) 
        6) question_2		(Question before playing): Wich is your degree
				of interest in economics or finance? (Low=0,
				Medium=1, High=2)
        7) question_3		(Question before playing): Have you ever
				operated in the stock market? (Yes=0, No=1)
        8) question_4		(Question before playing): Do you think that
				you can predict financial markets? (Yes=0,
				No=1, I don't know=2)
        9) question_5		(Question before playing): Do you think that
				experts can predict financial markets? (Yes=0,
				No=1, I don't know=2)
        10) question_6		(Question before playing): If you had the same
				information than the experts, then do you think
				that you can predict financial markets? (Yes=0,
				No=1, I don't know=2)
        11) question_7		(Question after playing): How have you found
				this experience? (Positive=0, Neutral=1,
				Negative=2)
        12) question_8		(Question after playing): Did you find the
				provided information useful for making
				decisions? (Yes=0, No=1, I don't know=2)
        13) question_9		(Question after playing): Which kind of
				information has been more useful? (Past 30 day
				price evolution = 0, Previous intraday price =
				1, Expert's opinion = 2, Arrows plot = 3, Other
				world markets = 4, None of them = 5)
        14) question_10		(Question after playing): How did you make your 
				decisions? (Completely by intuition = 0, More
				by intuition than by information = 1, Same
				degree of intuiton and information = 3, More by
				information than by intuition = 4, Completely 
				by information = 5, I don't know = 6)
        15) question_11		(Question before playing, same than
				Question_6): If you had the same information
				than the experts, then do you think that you
				can predict financial markets? (Yes=0, No=1, I
				don't know=2)
        16) score		Score or money at the end of the game.
        17) 1st_scenario_id	Corresponding game ID of first scenario.
        18) 2nd_scenario_id	Corresponding game ID of second scenario.
        19) 3rd_scenario_id	Corresponding game ID of third scenario.
        20) 4th_scenario_id	Corresponding game ID of fourth scenario.
        21) init_time		Time when user started playing.
        22) end_time		Time when user finished playing.
        23) finished		Flag users that have finished all steps.

    - Game: This table shows the information about games (sets of 25 rounds).
      The table is composed in the following fields:
        1) id			Game ID.
        2) scenario		First character indicates scenario while the
				second Intervention (I) or Control (C) groups.	
        3) series		Series ID played.
        4) completed		Tag for completed games (1). 
        5) correct_answers	Number of correct guesses.
        6) errors		Number of errors.
        7) end_date		Ending time.

    - Rounds: This table shows the information about rounds. The table is
      composed in the following fields:
        1) id				Round ID.
        2) game				Game ID.
        3) round			Number of round of the same game.
        4) user				User ID.
	5) scenario			First character indicates scenario
					while the second Intervention (I) or
					Control (C) groups.
        6) round_time			Total round time. 				
        7) decision			Decision Up (1), No answer(0) or
					Down(-1).
        8) result			Correct (1) or incorrect (-1).
        9) information_consulted	Number of information panels consulted.
        10) clicks			Number of clicks on buttons.
        11) info_daily_price_time	Time spended in the general screen.
        12) info_5days_average_time	Time spended in the 5 days average
					screen.
        13) info_30days_average_time	Time spended in the 30 days average
					screen.
        14) info_intraday_time		Time spended in the previous intraday
					price screen.
        15) info_expert_time		Time spended in the expert's screen.
        16) info_arrows_time		Time spended in arrows screen.
        17) info_world_markets_time	Time spended in world markets screen.

