Description

Part 2: Essentials of Choice Hypothesis Great choices : in light of rationale consider all accessible information and conceivable choices utilize a quantitative approach Awful choices : not in view of rationale don't consider all accessible information and conceivable options

Transcripts

Section 2: Fundamentals of Decision Theory

Good choices : taking into account rationale consider all accessible information and conceivable options utilize a quantitative approach Bad choices : not taking into account rationale don\'t consider all accessible information and conceivable choices don\'t utilize a quantitative methodology Decision Theory âan diagnostic and methodical way to deal with the investigation of choice makingâ A decent choice may every so often bring about a startling result; it is still a decent choice if settled on legitimately An awful choice may once in a while result in a decent result on the off chance that you are fortunate; it is still an awful choice

Steps in Decision Theory 1. Plainly characterize the current issue 2. List the conceivable options 3. Distinguish the conceivable results 4. List the result or benefit 5. Select one of the choice hypothesis models 6. Apply the model and settle on your choice

The Thompson Lumber Company Step 1: Clearly characterize the issue The Thompson Lumber Co. must choose whether or not to extend its product manufacturing so as to offer and promoting another item, lawn stockpiling sheds Step 2: List the conceivable choices elective : âa game-plan or method that may be picked by the choice makerâ (1) Construct a huge plant to fabricate the sheds (2) Construct a little plant (3) Do nothing

The Thompson Lumber Company Step 3: Identify the results (1) The business sector for capacity sheds could be great appeal (2) The business sector for capacity sheds could be unfavorable low request condition of nature : âan result over which the chief has practically zero controlâ

The Thompson Lumber Company Step 4: List the conceivable adjustments A result for every conceivable blend of options and conditions of nature Conditional qualities : âpayoff relies on the option and the condition of nature â with an ideal market: a huge plant creates a net benefit of $200,000 a little plant delivers a net benefit of $100,000 no plant creates a net benefit of $0 with an unfavorable business sector: an expansive plant creates a net loss of $180,000 a little plant creates a net loss of $20,000 no plant delivers a net benefit of $0

Payoff tables A method for sorting out a choice circumstance, including the settlements from distinctive circumstances given the conceivable conditions of nature Each choice, 1 or 2, results in a result, or result, for the specific condition of nature that happens later on May be conceivable to relegate probabilities to the conditions of nature to help in selecting the best result

The Thompson Lumber Company

The Thompson Lumber Company Steps 5/6: Select a proper model and apply it Model determination relies on upon the working environment and level of instability

Decision Making Environments Decision settling on under sureness Decision settling on under danger Decision making under vulnerability

Decision Making Under Certainty Decision creators know with assurance the outcomes of each choice option Always pick the option that outcomes in the best conceivable result

Decision Making Under Risk Decision creators know the likelihood of event for every conceivable result Attempt to augment the normal result Criteria for choice models in this environment: Maximization of expected fiscal worth Minimization of expected misfortune

Expected Monetary Value (EMV) EMV : âthe likelihood weighted whole of conceivable adjustments for each alternativeâ Requires a result table with contingent settlements and likelihood appraisals for all conditions of nature EMV(alternative i) = (payoff of first condition of nature) X (likelihood of first condition of nature) + (result of second condition of nature) X (likelihood of second condition of nature) + . . . + (result of last condition of nature) X (likelihood of last condition of nature)

Build the little plant The Thompson Lumber Company Suppose that the likelihood of a positive business sector is the very same as the likelihood of an unfavorable business sector. Which option would give the best EMV? EMV(large plant) = (0.5)($200,000) + (0.5)(- $180,000) = $10,000 EMV(small plant) = EMV(no plant) = $40,000 $0 (0.5)($100,000) + (0.5)(- $-20,000) = $40,000 (0.5)($0) + (0.5)($0) = $0

Expected Value of Perfect Information (EVPI) It might be conceivable to buy extra data about future occasions and in this manner settle on a superior choice Thompson Lumber Co. could enlist a financial expert to break down the economy with a specific end goal to all the more precisely figure out which monetary condition will happen later on How important would this data be?

EVPI Computation Look first at the choices under every condition of nature If data was accessible that flawlessly anticipated which condition of nature was going to happen, the best choice for that condition of nature could be made expected quality with immaculate data (EV w/PI): âthe expected or normal return on the off chance that we have impeccable data before a choice must be madeâ

EVPI Computation Perfect data changes environment from choice settling on under danger to choice making with conviction Build the extensive plant on the off chance that you know without a doubt that a good market will win Do nothing on the off chance that you know without a doubt that an unfavorable business sector will win

EVPI Computation Even however consummate data empowers Thompson Lumber Co. to settle on the right speculation choice, every condition of nature happens just a sure parcel of the time An ideal business sector happens half of the time and an unfavorable business sector happens half of the time EV w/PI ascertained by picking the best option for every condition of nature and increasing its result times the likelihood of event of the condition of nature

EVPI Computation EV w/PI = (best result for first condition of nature) X (likelihood of first condition of nature) + (best result for second condition of nature) X (likelihood of second condition of nature) EV w/PI = ($200,000)(0.5) + ($0)(0.5) = $100,000

EVPI Computation Thompson Lumber Co. would be silly to pay more for this data than the additional benefit that would be picked up from having it EVPI : âthe most extreme sum a leader would pay for extra data bringing about a choice superior to anything one made without impeccable data â EVPI is the normal result with immaculate data less the normal result without flawless data EVPI = EV w/PI - EMV EVPI = $100,000 - $40,000 = $60,000

Using EVPI of $60,000 is the greatest sum that Thompson Lumber Co. should pay to buy impeccable data from a source, for example, a business analyst âPerfectâ data is to a great degree uncommon A financial specialist commonly would be eager to pay some sum not exactly $60,000, contingent upon how solid the data is seen to be

Opportunity Loss An option way to deal with augmenting EMV is to minimize expected open door misfortune (EOL) Opportunity misfortune (lament) : âthe distinction between the ideal result and the genuine result receivedâ EOL is processed by developing an open door misfortune table and figuring EOL for every option

Opportunity Loss Table Opportunity misfortune (lament) for any condition of nature is computed by subtracting every result in the segment from the best result in the same section

Expected Opportunity Loss Closely identified with EMV : âthe likelihood weighted total of conceivable adjustments for each alternativeâ EOL : âthe likelihood weighted aggregate of conceivable second thoughts for each alternativeâ EOL(alternative i) = (regret for first condition of nature) X (likelihood of first condition of nature) + (lament for second condition of nature) X (likelihood of second condition of nature) + . . . + (lament for last condition of nature) X (likelihood of last condition of nature)

Minimum EOL EOL(large plant) = ? EOL(small plant) = ? EOL(no plant) = ?

Summary of Results Both criteria prescribed the same choice Not a fortuitous event; these two routines dependably bring about the same choice Repetitious to apply both techniques to a choice circumstance EV w/PI = $100,000 EMV = $ 40,000 EVPI = $ 60,000 = least EOL

Another Example A financial specialist is going to buy one of three sorts of land: a condo building, an office building, or a distribution center. The two future conditions of nature that will decide the amount of benefit the financial specialist will make are either great monetary conditions or awful monetary conditions. The benefits that will come about because of every choice given these two conditions of nature are outlined beneath:

EMV

EVPI EV w/PI = ? EVPI = ?

EOL

Decision Making Under Uncertainty When probabilities for the conceivable conditions of nature can be evaluated, EMV or EOL choice criteria are suitable When probabilities for the conceivable conditions of nature can not be surveyed, or can\'t be surveyed with certainty, other choice making criteria are obliged A circumstance known as choice settling on under instability Decision criteria include: Maximax Maximin Equal probability Criterion of authenticity Minimax lament

Maximax Criterion âGo for the Goldâ Select the choice that outcomes in the greatest of the most extreme settlements An exceptionally hopeful choice foundation Decision producer accept that the most good condition of nature for every choice option will happen

Maximax Thompson Lumber Co. accept that the most good condition of nature happens for every choice option Select the greatest result for every choice All three maximums happen if a positive economy wins (a tie if there should be an occurrence of no plant) Select the most extreme of the maximums Max