The first theory we looked at is the game theory which is a situation that has maultiple decision makers. Every one palying the game will in turn have an affect on the other person playing.
Within game theory there are such situations called play off's where the outcome of decision makers will have a pay off for each of the individuals palying.
Nash Equilibrium
The nash is achieved when both players have moved into positions where they have recieved the highest value possible to them.
With the collom and Row game players make there decisions from the expectations from what they believe the other player os going to make.
With the collom and row game for the nash to be achieved the row player would have to move low and the collom player would have to move right as this would then give the figure that both palyers could gain at there highest level. In this game both palyers are tyinging to maximise there play off's by working out the other players expectations. Row players can only move up and down ans collom players can only move left to right.
Dominance
A game that is used to show dominance bewteen individuals is the husband and wife game. Here a husband and wife have murdered an individual and only by working together will they be able to minimise the number of years they would recieve in jail as punishment.
As if one decide to confess and the other lies and vise verser this will leave either the husband or wife serving more years than the other. In there case here the Nash would be mutal confession as neither of them can be sure what the other will say confession will leave them serving the least amount of years as the lier in turn recieves more than double what they would recieve provided they confessed
Another game that can be used to demonstrate dominance is the game known as chicken where two people would drive into a dangerous situation and the fist to bail would be the looser. Here the only advantage each player has over one another would be to try and intimidate the other player or to try and trick them of there actuall intended movements. The nash here would be that one player swerves and the other player continues straight and vise verser.
PY3026C JUDGEMENT AND DECIDE
Tuesday, 7 December 2010
Thursday, 18 November 2010
Gain-Loss Framing and Choice
Gain and loss is all down to perception, as as a person who may perceive something entirely different than another individual may e.g a glass being half empty or a glass being half full. depending on how it is looked at will depend if a person takes it as a loss or a gain. According to the prospect theory out outcomes are experienced as gains or loss according the the individuals subjective reference point.
Subject value is said to be a concave function of utility and a steeper function in the domain of loss's.
Framing is said what can alter an individuals choice making decision, as framing is controlled by the formulation of the problem and partly by the individuals character as well as there habits likes and dislikes.
The effects of framing are mainly assessed by using two different types of criteria intergroup difference criterion and reference distribution Intergroup examines if there's a difference between the two relevant framing conditions. With reference distribution here it is required that the difference between framing is looked at and analysed .
Subject value is said to be a concave function of utility and a steeper function in the domain of loss's.
Framing is said what can alter an individuals choice making decision, as framing is controlled by the formulation of the problem and partly by the individuals character as well as there habits likes and dislikes.
The effects of framing are mainly assessed by using two different types of criteria intergroup difference criterion and reference distribution Intergroup examines if there's a difference between the two relevant framing conditions. With reference distribution here it is required that the difference between framing is looked at and analysed .
Tuesday, 9 November 2010
Prospect Theory
Prospect theory consist of two stages the editing and evaluation stage
Editing stage for the first stage is to simplify the problem and then look a solution.
Once you've found solution then you go move the decoding stage which is based on gains and loss due to your expectations.
Then you move to the combination stage which looks at the probability which an event is likely to occur.
After this comes to the segmentation stage here is where something people might do when separating problems for their outcomes.
The last stage in the editing stage is the councilation stage which is all about dismissing the common values, any values that are the same are not taken into account when counciling out which one to choose.
Then comes the evaluation stage
This stage has two components the value function and the weighting function.
In the value function is all to do with your expectations, if the values don't meet the expectations then it is perceived as a loss. Also here you more likely to become less sensitive to gains as the further you get from your expectations.
The next stage in the evaluation is the weighting function with this part of the evaluation stage the likely hood of things occurring here is due to sensitivity to changes in probabilities. The moment something is unlikely to occur it has a big impact on the person. here also small probabilities are overweighted bigger than they actually are
Editing stage for the first stage is to simplify the problem and then look a solution.
Once you've found solution then you go move the decoding stage which is based on gains and loss due to your expectations.
Then you move to the combination stage which looks at the probability which an event is likely to occur.
After this comes to the segmentation stage here is where something people might do when separating problems for their outcomes.
The last stage in the editing stage is the councilation stage which is all about dismissing the common values, any values that are the same are not taken into account when counciling out which one to choose.
Then comes the evaluation stage
This stage has two components the value function and the weighting function.
In the value function is all to do with your expectations, if the values don't meet the expectations then it is perceived as a loss. Also here you more likely to become less sensitive to gains as the further you get from your expectations.
The next stage in the evaluation is the weighting function with this part of the evaluation stage the likely hood of things occurring here is due to sensitivity to changes in probabilities. The moment something is unlikely to occur it has a big impact on the person. here also small probabilities are overweighted bigger than they actually are
Thursday, 28 October 2010
Tuesday, 19 October 2010
Take the best for get the rest (Algorithm)
The take the best Algorithm is a model used for fast and quick analysis with limited knowledge with the view of take the best and for get the rest. this module has 5steps is use's before it in turn comes to its final decision.
The first step that the module adopts is to choose any object that is recognised when selecting out of the two. Where as if neither of the two objects are recognised then choose randomly, but if both of the objects are recognised then proceed onto step two.
The second step the model adopts once you have both of the objects then you retrieve the cue vales of the objects with the highest ranking of cue memory.
The third step of the Algorithm is to decide if the cue discriminates or not. Providing that both values are the same the cue isn't known to discriminate unless one has a positive value and the other doesn't.
Step four of the model is to then search for the discrimination as if the cue doesn't discriminate the your advised to go back to step two until the discrimination is found. If the cue is found to discriminate then proceed on to the final step
The final step used by the model is to then choose the cue that has a positive value, but if no cue discriminates then your advised to choose randomly.
For example if we are trying to figure out which city is larger city a or b for the first step we would have to have both cities recognised. Step two would be then to search for the best cue results with a positive and negative value. Step three is then to see if the cue value then discriminates if it dose then you would move onto step four where the search would stop and from here the person would make the assumption that city a rather than b is larger than a which is the step Five as the final decision has been made.
The first step that the module adopts is to choose any object that is recognised when selecting out of the two. Where as if neither of the two objects are recognised then choose randomly, but if both of the objects are recognised then proceed onto step two.
The second step the model adopts once you have both of the objects then you retrieve the cue vales of the objects with the highest ranking of cue memory.
The third step of the Algorithm is to decide if the cue discriminates or not. Providing that both values are the same the cue isn't known to discriminate unless one has a positive value and the other doesn't.
Step four of the model is to then search for the discrimination as if the cue doesn't discriminate the your advised to go back to step two until the discrimination is found. If the cue is found to discriminate then proceed on to the final step
The final step used by the model is to then choose the cue that has a positive value, but if no cue discriminates then your advised to choose randomly.
For example if we are trying to figure out which city is larger city a or b for the first step we would have to have both cities recognised. Step two would be then to search for the best cue results with a positive and negative value. Step three is then to see if the cue value then discriminates if it dose then you would move onto step four where the search would stop and from here the person would make the assumption that city a rather than b is larger than a which is the step Five as the final decision has been made.
Subscribe to:
Posts (Atom)