1 / 32

Random Sampling - Random Samples

Random Sampling - Random Samples. Why do we need Random Samples?. Many business applications -We will have a random variable X such that the probability distribution & expected value is unknown -The only way to make use of probability is to estimate E(X) and if possible F x or f x

Download Presentation

Random Sampling - Random Samples

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Random Sampling- Random Samples

  2. Why do we need Random Samples? • Many business applications -We will have a random variable X such that the probability distribution & expected value is unknown -The only way to make use of probability is to estimate E(X) and if possible Fx or fx -This can be done with random sampling

  3. Random Samples X come from some random process. x results from a trial of the process (observation of X ) a set {x1, x2, , xn} of nindependent observations of the same random variable X is called a random sample of size n.

  4. Sample Mean • What does a random sample tell us about a random variable? • Consider random sample set {x1, x2, , xn} SAMPLE MEAN

  5. Example

  6. Important • can be used as an estimate of the parameter E(X). • In general, the larger the sample size n, the better will be the estimate

  7. APPROXIMATING MASS AND DENSITY FUNCTIONS If we have a large enough sample, we can group the data and form a histogram that approximates • the probability mass function (for a finite random variable) or • the probability density function (for a continuous random variable).

  8. APPROXIMATING MASS FUNCTIONS-Discrete random variables We have used bins of width 1 and have plotted relative frequencies. The relative frequency of each value of X in the sample gives an estimate for the probability that X will assume that value. Hence, the relative frequency of a value x in the sample approximates P(X = x) = fX(x) [p.m.f - Discrete random variables]

  9. APPROXIMATING DENSITY FUNCTION-Continuous random variables • Recall that the p.d.f can be used to find probabilities P(a X b) is equal to the area under the curve of the p.d.f over the interval [a,b] • If we want to use a histogram approximate the p.d.f then Relative frequency of a bin= Area of the corresponding rectangle

  10. Important • But we know Area of rectangle=width x height But Relative frequency of a bin= Area of the corresponding rectangle • Now the area of each rectangle represents the probability • Now we must plot the adjusted relative frequencies against the mid points of the bins

  11. Approximating the p.d.f. (Disney) Histogram function is used for normalized ratios( Rnorm) Since the width is 0.03 0.0024/0.03 (0.73+0.76)/2

  12. Approximating the p.d.f. (Disney) Height Normalized ratios

  13. Approximating the p.d.f. (Disney) Height Normalized ratios

  14. Stock Option Pricing Probability, Mathematics, Tests, Homework, Computers on the project Random Samples, Focus How can random sampling help us price a stock option? Recall that the continuous random variable Rnorm gives the normalized ratio of weekly closing prices on Walt Disney stock. The sheet Sample of the file Option Focus.xls computes 417 values of this normalized ratio from our 417 weekly closing ratios. We will assume that these are independent observations of Rnorm, that constitute a random sample of size 417 for Rnorm. Class Project  T I  C Option Focus.xls (material continues)

  15. Stock Option Pricing Probability, Mathematics, Tests, Homework, Computers on the project Random, Focus The sample mean, 1.0007695, of these observations is computed in the sheet Sample of Option Focus.xls. This is the same as our estimate for the ratio, Rrf, which corresponds to the weekly risk-free interest rate. Since we constructed the normalized ratios to make this true, we have a check on the correctness of our work. The HISTOGRAM function is used to group the sample data for Rnorm, and a plot is created with the total area for all of the bars being equal to 1. This produces the bar graph, shown in the sheet Sample, which approximates the p.d.f., fnorm, of Rnorm. Connecting the midpoints at the tops of the bars produces the line graph approximation for fnorm that is shown in Sample. Our plots, which are also shown on the next page, give a visual indication for the volatility of Disney stock over the past 8 years. Class Project  T I  C Option Focus.xls (material continues)

  16. Stock Option Pricing Probability, Mathematics, Tests, Homework, Computers on the project Random, Focus WHAT SHOULD YOU DO? Each team should now plot an approximation of the probability density function for the normalized ratios of weekly closing prices for its particular stock data and should find the sample mean of the normalized ratios. Class Project  T I C Option Focus.xls (material ends)

  17. Important • Check class website • Where you should be with the Project • Normalize the ratios of closing prices • Create a histogram of normalized ratios • Read the Requirements for the Project 2 written report • Write a draft of the Discussion of Options section of written report What’s ahead? • Incorporate your simulation results for RANDBETWEEN in the Discussion of Simulation section • Decide which assumption that you will discuss in more detail in the written report

  18. fX X Expected Value To consider the expected value (or mean) of a continuous random variable, we can use the probability density function(p.d.f) to give us a geometric interpretation. The expected value perfectly “balances” the area to its right with the area to its left.

  19. Estimate the E(X) for the continuous random variable X, whose p.d.f., fX is shown below. Exercise 33 Hw11 E(X)=4 (balances the area to its right with the area to its left

  20. Bootstrapping • collect a smaller sample of data points, then use a computer to simulate a much larger set.

  21. Computer Simulations • Why do simulations? • Cost benefits • Time constraints • Availability of data • Which Excel functions will we need? • RANDBETWEEN() – discrete case - VLOOKUP() • IF()

  22. Simulation, Integers Read the description of the function from the Excel menu. RANDBETWEEN FUNCTION

  23. Simulation, Integers This process uses the function VLOOKUP, which is found in the Lookup & Reference submenu of the Function Wizard. Read the description of the function from the Excel menu. VLOOKUP FUNCTION

  24. Simulation, Integers Simulation. Integers: page 6 Value in the leftmost column of table Location of table Number of the column where value is to be found For our purposes, this will usually be left blank. Phone Log.xls  T I  C (material continues)

  25. The IF function is found under the Logical submenu of the Function Wizard.

  26. Read the description of the function from the Excel menu. If B12 is less than or equal to 0.5, the function returns an H. If B12 is not less than or equal to 0.5, a T is returned.Note that the desired text must be specified in quotation marks. For the stock option project we do not need quotation marks-will show later in excel

  27. More on Project Focus • r1 -a normalized ratio of the adjusted closing price at the end of the first week (will be selected randomly) • initial price - $21.8700 • r1 is a value of Rnorm that might have occurred for the first week.

  28. Likewise, let r2, r3, , r20 be normalized ratios of adjusted closing prices for weeks 2, 3, , 20 of the option. • It is stated in the project description that observations of R are all independent. Thus, the normalized ratios r1, r2, r3 ,  , and r20 are all independent observations of Rnorm

  29. random observation for the ratio r1r2r3r20 • cnorm = $21.8700r1r2r3r20 • cnorm is a observation of normalized closing price, Cnorm. • We conclude that any set of 20 observations of R determines a set of 20 observations of Rnorm and, therefore, an observation of Cnorm.

  30. FV - the continuous random variable giving the per share value of our Walt Disney call at the end of twenty weeks, based on the normalized closing price. • s0 - strike price of $23 • Cnorm takes on a value cnorm,

  31. Recall • if closing stock price strike price • Final value of Call option (Intrinsic Value of a Call) =Maximum of 0 and C – S • Case 1. FV = cnorm s0 if cnorm s0, • Case2. FV=0 if cnorm < s0.

  32. 20 observations of Rnorm determine an observation of Cnorm, which, together with the strike price, determines an observation of FV. • PV - the continuous random variable giving the present value of the Disney option on January 11, 2002. • The present value for a value of FV is an observation of PV. Our price for the present value of the option is the expected value of PV.

More Related