Вы находитесь на странице: 1из 11

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

CHAPTER 6 (SIX)

Decision trees and influence diagrams can be extremely useful in helping people to gain an understanding of the structure of the problems which confront them. Decision problems are multi-stage in character when the choice of a given option may result in circumstances which will require yet another decision to be made. Decision trees can serve a number of purposes when complex multi-stage problems are encountered. They can help a decision maker to develop a clear view of the structure of a problem and make it easier to determine the possible scenarios which can result if a particular course of action is chosen. This can lead to creative thinking and the generation of options which were not previously being considered. Decision trees can also help a decision maker to judge the nature of the information which needs to be gathered in order to tackle a problem and, because they are generally easy to understand, they can be an excellent medium for communicating one persons perception of a problem to other individuals. Influence diagrams offer an alternative way of structuring a complex decision problem and some analysts find that people relate to them much more easily. Influence diagrams can be converted to decision trees and we will therefore regard them in this chapter as a method for eliciting decision trees.

Constructing a decision tree A square is used to represent a decision node and, because each branch emanating from this node presents an option, the decision maker can choose which branch to follow. A circle, on the other hand, is used to represent a chance node. The branches emanating from a circle are therefore labeled with probabilities which represent the decision makers estimate of the probability that a particular branch will be followed.

SUFIAN (29112017) X-47

Page 1

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

Determining the optimal policy It can be seen that our decision tree consists of a set of policies. A policy is a plan of action stating which option is to be chosen at each decision node that might be reached under that policy. The technique for determining the optimal policy in a decision tree is known as the rollback method. To apply this method, we analyze the tree from right to left by considering the later decisions first. It can be seen that the rollback method allows a complex decision problem to be analyzed as a series of smaller decision problems.

Decision trees involving continuous probability distributions The method is based on earlier work by Pearson and Tukey and requires three estimates to be made by the decision maker: (i) The value in the distribution which has a 95% chance of being exceeded. This value is allocated a probability of 0.185. (ii) The value in the distribution which has a 50% chance of being exceeded. This value is allocated a probability of 0.63. (iii) The value in the distribution which has only a 5% chance of being exceeded. This value is also allocated a probability of 0.185. Nevertheless, in general, there are clear advantages in using this approximation. Above all, it is simple and each distribution requires only three estimates to be made which has the obvious effect of reducing the decision makers judgmental task.

Assessment of decision structure It is really a matter of the decision analysts judgment as to whether the elicited tree is a fair representation of the decision makers decision problem. Once a structure is agreed then the computation of expected utility is fairly straightforward. The figure below presents a description

SUFIAN (29112017) X-47

Page 2

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

of the typical phases in a decision analysis of a problem that the decision maker wishes to resolve with help of the practitioner of decision analysis the decision analyst.

Eliciting decision tree representations What methods have been developed to help elicit decision tree representations from decision makers? One major method, much favored by some decision analysts, is that of influence diagrams which are designed to summarize the dependencies that are seen to exist among events and acts within a decision.

SUFIAN (29112017) X-47

Page 3

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

In addition, influence diagrams are more easily revised and altered as the decision maker iterates with the decision analyst. Decision trees, because of their strict temporal ordering of acts and events, need completely respecifying when additional acts and events are inserted into preliminary representations. One step-by-step procedure for turning an influence diagram into a decision tree is as follows: (1) Identify a node with no arrows pointing into it (since there can be no loops at least one node will be such). (2) If there is a choice between a decision node and an event node, choose the decision node. (3) Place the node at the beginning of the tree and remove the node from the influence diagram. (4) For the now-reduced diagram, choose another node with no arrows pointing into it. If there is a choice a decision node should be chosen. (5) Place this node next in the tree and remove it from the influence diagram. (6) Repeat the above procedure until all the nodes have been removed from the influence diagram.

Finally, we analyzed the process of generating decision tree representation of decision problems and advocated the influence diagram as a key technique to facilitate decision structuring.

SUFIAN (29112017) X-47

Page 4

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

CHAPTER 8 (EIGHT)

Bayes theorem Bayes theorem will be used as a normative tool, telling us how we should revise our probability assessments when new information becomes available. In Bayes theorem an initial probability estimate is known as a prior probability. The steps in the process are summarized below: (1) Construct a tree with branches representing all the possible events which can occur and write the prior probabilities for these events on the branches. (2) Extend the tree by attaching to each branch a new branch which represents the new information which you have obtained. On each branch write the conditional probability of obtaining this information given the circumstance represented by the preceding branch. (3) Obtain the joint probabilities by multiplying each prior probability by the conditional probability which follows it on the tree. (4) Sum the joint probabilities. (5) Divide the appropriate joint probability by the sum of the joint probabilities to obtain the required posterior probability.

The effect of new information on the revision of probability judgments It is interesting to explore the relative influence which prior probabilities and new information have on the resulting posterior probabilities. At the extreme, if your prior probability of an event occurring is zero then the posterior probability will also be zero. Whatever new information you receive, no matter how reliable it is, you will still refuse to accept that the event is possible. In general, assigning prior probabilities of zero or one is unwise. Ironically, if the new information has less than a 0.5 chance of being reliable its result is of interest and the more unreliable it is, the greater the effect it will have on the prior probability.

SUFIAN (29112017) X-47

Page 5

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

Applying Bayes theorem to a decision problem This simply involves the use of the posterior probabilities, rather than the prior probabilities, in the decision model.

Assessing the value of new information New information can remove or reduce the uncertainty involved in a decision and thereby increase the expected payoff. However, in many circumstances it may be expensive to obtain information since it might involve, for example, the use of scientific tests, the engagement of the services of a consultant or the need to carry out a market research survey.

The expected value of perfect information In many decision situations it is not possible to obtain perfectly reliable information, but nevertheless the concept of the expected value of perfect information (EVPI) can still be useful. We emphasize that our calculations are based on the assumption that the decision maker is risk neutral. If the manager is risk averse or risk seeking or if he also has non-monetary objectives then it may be worth him paying more or less than this amount.

The expected value of imperfect information As with the expected value of perfect information, we will need to consider the possible indications the test will give, what the probabilities of these indications are and the decision the manager should take in the light of a given indication. A summary of the main stages is given below: (1) Determine the course of action which would be chosen using only the prior probabilities and record the expected payoff of this course of action; (2) Identify the possible indications which the new information can give;

SUFIAN (29112017) X-47

Page 6

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

(3) For each indication: (a) Determine the probability that this indication will occur; (b) Use Bayes theorem to revise the probabilities in the light of this indication; (c) Determine the best course of action in the light of this indication (i.e. using the posterior probabilities) and the expected payoff of this course of action; (4) Multiply the probability of each indication occurring by the expected payoff of the course of action which should be taken if that indication occurs and sum the resulting products. This will give the expected payoff with imperfect information; (5) The expected value of the imperfect information is equal to the expected payoff with imperfect information (derived in stage 4) less the expected payoff of the course of action which would be selected using the prior probabilities There is an alternative way of looking at the value of information. New information can be regarded as being of no value if you would still make the same decision regardless of what the information told you.

Practical considerations Clearly, it is easier to identify the expected value of perfect as opposed to imperfect information, and we recommend that, in general, calculating the EVPI should be the first step in any information-evaluation exercise. The EVPI can act as a useful screen, since some sources of information may prove to be too expensive, even if they were to offer perfectly reliable data, which is unlikely. In most cases, however, the assessment of the reliability of the information will ultimately be based on the decision makers subjective judgment. The inputs required by the decision maker are: (1) the maximum loss which will need to be incurred before the product is removed from the market, (2) the probability of incurring this loss if the product is introduced, (3) the probability that the product will be successful, if introduced and (4) the probability that the market research will accurately predict the true state of the market.

SUFIAN (29112017) X-47

Page 7

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

CHAPTER 9 (NINE)

Most of the methods that are designed to help people to make decisions under conditions of uncertainty require estimates of probabilities for the possible outcomes of the decision. Because of the unavailability of suitable statistical data, these probabilities will usually be subjective estimates based on human judgment.

Heuristics and biases Much of the research on the quality of human judgment of probability has stemmed from the work of Tversky and Kahneman. The central theme of Tversky and Kahnemans work is that people use rules of thumb or heuristics to cope with the complexities of making estimates of probabilities. Three main heuristics identified by Tversky and Kahneman are (i) availability, (ii) representativeness and (iii) anchoring and adjustment.

The availability heuristic People using the availability heuristic therefore judge the probability of the occurrence of events by how easily these events are brought to mind. Availability can be a reliable heuristic. Frequently occurring events are usually easier to recall so the higher probabilities estimates for them should be reliable. For example, some events are easily recalled precisely because they are unusual and rare. This can lead to biased estimates. Biases associated with the availability heuristic 1. When ease of recall is not associated with probability Easily recalled events are not necessarily highly probable. 2. Ease of imagination is not related to probability Easily imagined events are not necessarily highly probable. 3. Illusory correlation

SUFIAN (29112017) X-47

Page 8

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

In decision analysis models, illusory correlation is of concern when conditional probabilities.

The representativeness heuristic If you use the representativeness heuristic you will answer questions by judging how representative the object, person or event is of the category or process. Biases associated with the representativeness heuristic 1. Ignoring base-rate frequencies 2. Expecting sequences of events to appear random 3. Expecting chance to be self-correcting 4. Ignoring regression to the mean 5. The conjunction fallacy

The anchoring and adjustment heuristic Judgment is widely used to make estimates of values such as how long a job will take to complete or what next months sales level will be. Often these estimates start with an initial value which is then adjusted to obtain the final estimate. Typical initial values might be how long the last job took to complete or this months level of sales. Unfortunately, the adjustment from these initial values is often insufficient; a phenomenon known as anchoring. Biases associated with anchoring and adjustment 1. Insufficient adjustment In decision making, anchoring can be a problem in the estimation of costs, payoffs and probabilities. Forecasts that are used in the decision process may be biased by forecasters anchoring on the current value and making insufficient adjustment for the effect of future conditions. 2. Overestimating the probability of conjunctive events

SUFIAN (29112017) X-47

Page 9

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

Typical examples might be: the main machine and the back-up machine both fail today or you get promoted and win the lottery and write a best-selling novel all within the next 12 months. Each of the individual events which might co-occur is called an elementary event. The tendency of people to overestimate probabilities for conjunctive events may therefore lead to unjustified optimism that the project will succeed. 3. Underestimating probabilities for disjunctive events When asked to estimate the probability of a disjunctive event it appears that, once again, people anchor on one of the elementary events. With disjunctive events this leads to a tendency to underestimate the probability. Since the estimation of risk often involves probability assessments for disjunctive events, this bias can be a serious concern. 4. Overconfidence It appears that overconfidence is most marked when people lack expertise or knowledge about the quantity they are estimating. People with expertise or relevant knowledge tend not to be overconfident, and they may even have a tendency to be under confident.

Other judgmental biases 1. Believing desirable outcomes are more probable Much research has suggested that people tend to view desirable outcomes as more probable than those which are undesirable. 2. Biased assessment of covariation Several researchers have found that a similar bias can occur when people are presented with tables showing the number of times that events either occurred or failed to occur together

Is human probability judgment really so poor? Over the last decade much of the work on heuristics and biases has been questioned. Most criticisms have centered on the fact that the research has largely been based on

SUFIAN (29112017) X-47

Page 10

MM 5010 Strategic Decision Making & Negotiation

MBA - ITB

inexperienced decision makers carrying out artificial tasks in psychological laboratories, rather than real-world decision makers making real decisions. 1. Subjects in studies may be unrepresentative of real decision makers 2. Laboratory tasks may be untypical of real-world problems 3. Tasks may be misunderstood by subjects 4. Subjects may be poorly motivated 5. Citation bias 6. Real world studies suggest better performance 7. People think in terms of frequencies not probabilities

A methodology for choosing how to develop a subjective probability assessment

SUFIAN (29112017) X-47

Page 11

Вам также может понравиться