MONTE CARLO VARIANCE REDUCTION METHODS

Size: px
Start display at page:

Download "MONTE CARLO VARIANCE REDUCTION METHODS"

Transcription

1 MONTE CARLO VARIANCE REDUCTION METHODS M. Ragheb /4/3. INTRODUCTION The questio arises of whether oe ca reduce the variace associated with the samplig of a radom variable? Ideed we ca, but we eed to be somehow sophisticated i samplig the radom variable. There is a wealth of methods ad techiques that allow us to reduce the variace: these are called variace reductio methods. We shall cosider some of them here. These methods are sometimes problem-depedet, but their usage i some cases would make possible the solutio of problems that would be otherwise itractable, because of the tremedous amout of computer time that would be eeded i a direct simulatio. This is the cause of the miscoceptio of the Mote Carlo method beig expesive. Ideed, if variace reductio methods are ot used i some cases, the umerical experimet may become more costly tha buildig a real physical experimet. The kowledge of variace reductio methods is thus a requiremet for the clever ad ecoomical use of the Mote Carlo method.. VARIANCE REDUCTION BY MODIFICATION OF THE SAMPLING SCHEME Every radom variable has a uderlyig variace associated with its probability desity fuctio. For istace, the radom variable: : Poits obtaied i throwig a sigle die : () The mea value ad the variace of this radom variable are: ( ) ( ) (3.5)

2

3 3. ANALOG, HIT OR MISS, OR POOR MAN'S SAMPLING We cosider variace reductio for the estimatio of simple itegrals sice the basic priciples ivolved geeralize quite easily to more complex problems, such as particle trasport ad fluid flow. Suppose we eed to estimate the itegral: with: f( x), whe x. Let us draw the curve: f ( x) dx (3) y f ( x), withi the uit square, as show i Fig.. We may write: where: gx ( ) =, if f ( x) y =, if f ( x) y. g( x, y) dy f ( x), (4) Simply stated, we accept ay sampled poits that lie below the curve for the mea estimatio process ad reject those that do ot. We the ca estimate θ as a double itegral: usig the estimator: g( x, y) dxdy, g i * g i, i (5) where: * is the umber of occasios o which f, i i i is a uiformly sampled y-coordiate poit over the iterval [,], is a uiformly sampled x-coordiate poit over the uit iterval, i i, i thus determies a poit uiformly distributed over the uit square.

4 I fact, we sample poits at radom i the uit square, ad cout the proportio of them which lie below the curve y f ( x). This is i retrospect the rejectio samplig method. Figure. Hit-or-Miss, Aalog, or Poor Ma's Mote Carlo estimatio for the estimatio of a itegral. As a umerical example, let us estimate the itegral: I x e dx (6) which has a exact aalytical value: I e x To have f ( x) whe x, we perform a simple scalig:

5 I e x e dx e e g( x) dx e ' (7) ad we the estimate the value of θ' from which we ca evaluate θ. A procedure usig the Aalog Mote Carlo Itegratio method is show i Fig..! aalog.f9! Aalog, Hit or Miss, or Poor Ma's Mote Carlo! Estimatio of the value of itegral usig Aalog Mote Carlo! I=Itegral e**x over the iterval [,] =.78888! M. Ragheb program aalog real mea real :: e= real :: trials=! Defie sampled fuctio; fuctio is scaled to lie withi the! uit square g(x) = exp(x)/e! Iitialize score score =.! Sample poits o uit square do i=,trials call radom(rr) x=rr f=g(x) call radom(rr) y=rr if(y.le.f)the score =score+. ed if ed do! Estimate mea value mea=score/trials! rescale the mea value mea = mea * e! Write results write(*,*) mea, trials ed Figure. Procedure for itegratio usig Aalog, Hit-or-miss, or Poor Ma s Mote Carlo. I the case of the hit ad miss Mote Carlo, the variace is that of the biomial distributio give by: p( p), p( p I our case, p is the umber of hits to the umber of trials ratio: * p Some computatioal results are show i Table. As a example to estimate the variace ad the stadard deviatio of the biomial distributio:

6 * score.7633 p trials ( p) p( p) p( p Table. Mea values ad Fractioal Stadard Deviatio usig Aalog, Hit-or-miss, or Poor Ma s Mote Carlo. fsd / , , It ca be oticed that a excessive umber of trials is required to get a low fractioal stadard deviatio. The stadard deviatio is that of the biomial distributio. 4. CRUDE MONTE CARLO SAMPLING If ξ, ξ, ξ3,, ξ are idepedet radom umbers uiformly distributed over the uit iterval [.], the the quatities: f i f( ) (8) i are idepedet radom variates with expectatio θ. Therefore, accordig to the Cetral Limit Theorem: f i f i (9) Table. Mea Values ad Stadard Deviatios usig Crude Mote Carlo. fsd / , ,

7 is a ubiased estimator of θ, ad its variace is: The stadard error of f is: f x f dx (). f From Table, it appears that Crude Mote Carlo does ot improve i term of the achieved error estimate over Hit or Miss Mote Carlo. I Fig. 3 we show how simple the procedure for crude Mote Carlo for the estimatio of the itegral is. I Fig. 4 the procedure is modified for the estimatio of the variace usig the covetioal estimatio formula. Figure 5 shows a alterative way of estimatig the variace usig the earlier derived recursive variace estimatio formula.! crude f9! Crude Mote Carlo! I=Itegral[e**x,,]=.78888! M. Ragheb program crude real mea real :: trials=! Defie sampled fuctio g(x) = exp(x)! Iitialize score score =.! Sample fuctio do i=,trials call radom(rr) score =score+g(rr) ed do! Estimate mea value mea=score/trials! Write results write(*,*) mea, trials ed Figure 3. Procedure for itegratio usig Crude Mote Carlo.! crude.f9! Crude Mote Carlo! Variace estimatio usig covetioal formula! I=Itegral[e**x,,]=.78888! M. Ragheb program crude real mea real :: trials=! Defie sampled fuctio g(x) = exp(x)! Iitialize scores score =. sum_squares =.! Sample fuctio do i=,trials

8 call radom(rr) s = g(rr) score =score + s sum_squares = sum_squares + s *s ed do! Estimate mea value, variace, stadard deviatio,! ad fractioal stadard deviatio mea=score/trials variace = sum_squares/trials - mea*mea variace = (trials/(trials-.)) * variace std_dev = sqrt (variace) std_error = std_dev/sqrt(trials) frac_sta_dev = std_error/mea! Write results write(*,*) mea, std_error, frac_sta_dev, trials ed Figure 4. Procedure for itegratio usig Crude Mote Carlo, with covetioal formula for the estimatio of the variace.! crude.f9! Crude Mote Carlo! Variace estimatio usig recursive formula! I=Itegral[e**x,,]=.78888! M. Ragheb program crude real mea_value, t, ta real :: trials=! Defie sampled fuctio g(x) = exp(x)! Iitialize variables t=. s_square_old =.! Sample fuctio do k=,trials xk=k call radom(rr) s = g(rr) if (k.ge.) the ta = t/(xk-.) & ed if t=t+s s_square_ew = (xk-.)/(xk-.)*s_square_old & + (ta-s)*(ta-s)/xk s_square_old = s_square_ew ed do! Estimate mea value, variace, stadard deviatio,! ad fractioal stadard deviatio mea_value = t/trials std_dev = sqrt (s_square_ew) std_error = std_dev/sqrt(trials) frac_sta_dev = std_error/mea_value! Write results write(*,*) mea_value, std_error, frac_sta_dev, trials ed Figure 5. Procedure for itegratio usig Crude Mote Carlo, with recursive formula for the estimatio of the variace. 5. CORRELATED SAMPLING OR CONTROL VARIATES The equatio:

9 f ( x) dx is broke ito two parts by addig the subtractio a easy fuctio ( x) that approximates the fuctio f(x): ( x) dx f ( x) ( x) dx which are itegrated separately, the first part by mathematical theory ad the secod part by crude Mote Carlo. This amouts to estimatig a correctio [f(x)-(x)] to the fuctio ( x), which approximates f(x). Deotig: ( x) dx, the used estimator would be: f ( x) ( x) ux ( ) () px ( ) with p(x) as a uiformly distributed probability desity fuctio. The followig choices for ( x) ca be cosidered: x ( x) x ( ) x x x 3 x x 3 x x x x x 4 x x 6 4 which are i fact successive approximatios to the itegrad e x. Samplig from these s ca be carried out by use of the rejectio method. The correspodig s are:

10 A procedure for the applicatio of the correlated sample tevhique is show i Fig. 6.! correlated f9! Correlated Samplig or Cotrol Variates Mote Carlo! Variace estimatio usig recursive formula! I=Itegral[e**x,,]=.78888! M. Ragheb program correlated real mea_value, t, ta real :: trials=5! Defie sampled fuctio g(x) = exp(x)! Defie cotrol variate fuctio f(x) =. + x! Aalytically itegrated value of f(x) phi =.5! Ope output file ope(44, file = 'radom_out')! Iitialize variables t=. s_square_old =.! Sample fuctio do k=,trials xk=k call radom(rr)! Evaluate cotrol variate s = (g(rr) - f(rr)) + phi if (k.ge.) the ta = t/(xk-.) & ed if t=t+s s_square_ew = (xk-.)/(xk-.)*s_square_old & + (ta-s)*(ta-s)/xk s_square_old = s_square_ew ed do! Estimate mea value, variace, stadard deviatio,! ad fractioal stadard deviatio mea_value = t/trials std_dev = sqrt (s_square_ew) std_error = std_dev/sqrt(trials) frac_sta_dev = std_error/mea_value! Write results write(*,) mea_value, std_error, frac_sta_dev, trials write(44,) mea_value, std_error, frac_sta_dev, trials format(x,'mea value =',e4.8,/,x,'stadard error =',e4.8,/, & & x,'fractioal stadard deviatio =',e4.8,/, & & ed x,'umber of trials =',e4.8) Figure 6. Procedure for the applicatio of the correlated samplig techique. A drastic decrease i variace compared with aalog ad crude samplig was detected by goig from a good to a better approximatio for f(x), some of the results beig show i the followig Table 3. Table 3. Mea Values ad Stadard Deviatios usig Correlated Samplig. fsd /

11 ( x) x ( ) x.738. x x x x 3 x x ,, x x x 4 x x IMPORTANCE SAMPLING We have: f ( x) dx f( x) gx ( ) g( x) dx f( x) dg ( x ) gx ( ) (3) for ay fuctios g ad G where: dg( x) g( x) dx ; ad if g is positive valued ad ormalized such that: G() g( x) dx. If η is a radom umber sampled from the distributio G, the: f ( ) E g ( ) (4) Thus, a equivalet estimator would be: f( x) vx ( ). ( x) (5) with the samplig carried out from the probability desity fuctio: ( x) px ( ).

12 ! importace f9! Importace Samplig Mote Carlo! Variace estimatio usig expoetial easy fuctio ad recursive formula! I=Itegral[e**x,,]=.78888! M. Ragheb program importace real mea_value, t, ta real :: alpha=.9 real :: trials=5! Defie sampled fuctio f(x) = exp(x)! Defie importace fuctio(ormalized) g(x) =(alpha/(exp(alpha)-.))*exp(alpha*x)! Ope output file ope(44, file = 'radom_out')! Iitialize variables t=. s_square_old =.! Sample fuctio do k=,trials xk=k! Sample Importace Fuctio g(x) call radom(rr) x=log(.+((exp(alpha)-.)*rr))/alpha! Evaluate importace estimator s = f(x)/g(x) if (k.ge.) the ta = t/(xk-.) s_square_ew = (xk-.)/(xk-.)*s_square_old & & + (ta-s)*(ta-s)/xk s_square_old = s_square_ew ed if t=t+s ed do! Estimate mea value, variace, stadard deviatio,! ad fractioal stadard deviatio mea_value = t/trials std_dev = sqrt (s_square_ew) std_error = std_dev/sqrt(trials) frac_sta_dev = std_error/mea_value! Write results write(*,) mea_value, std_error, frac_sta_dev, trials write(44,) mea_value, std_error, frac_sta_dev, trials format(x,'mea value =',e4.8,/,x,'stadard error =',e4.8,/, & & x,'fractioal stadard deviatio =',e4.8,/, & & x,'umber of trials =',e4.8) ed Figure 7. Procedure for the applicatio of the Importace Samplig techique usig the expoetial fuctio. Results of computatios are show i the followig Table 4 for the cases: ( x) x ( ) x x x 3 x x 3 x x 6

13 3 4 x x x 4 x x 6 4 as successive approximatios of the itegral. The rejectio method was used to sample these fuctios. Better results tha for correlated samplig were obtaied. Table 4. Mea Values ad Stadard Deviatios for Importace Samplig. ( x) x ( ) x x x x x 3 x x 6 / fsd , x x x 4 x x WEIGHTED UNIFORM SAMPLING We use the estimator: w,,..., 3. i i g i i N D (6) where the ξi s are idepedet radom umbers with idetical probability desity fuctio p(s): ad: f ( s) g( s) where ps ( ) # p( s) where p ( s) () s () s where ps ( ) # ps () where p ( s) The followig approximatig fuctios were used: ( x) x

14 ( ) x x x 3 x x 3 x x x x x 4 x x 6 4 Results of computatios are show i Table 5 below. Table 5. Mea Values ad Stadard Deviatios for Weighted Uiform Samplig. ( x) x ( ) x x x x x x x / fsd , x x x 4 x x ANTITHETIC VARIATES I the Atithetic Variates method, variace reductio is achieved by symmetrizatio of the itegral through a group trasformatio. A estimator that ca be used is: t f ( ) f ( ) (7) which amouts to a average of the fuctio f() ad its mirror image f(-). As show i Fig., less variatio from the mea ca be expected from the estimator t tha by usig f() or f(-) aloe.

15 Figure 8. Variace reductio through the use of the Atithetic Variates estimator: t f f 9. STRATIFICATION Aother estimator based o the Atithetic Variates priciple uses stratificatio of the itegratio iterval such as: t f f f f 4 (8)! atithetic f9! Atithetic Variate Mote Carlo! Variace estimatio usig recursive formula! I=Itegral[e**x,,]=.78888! M. Ragheb program importace real mea_value, t, ta

16 real :: trials=5! Defie sampled fuctio f(x) = exp(x)! Ope output file ope(44, file = 'radom_out')! Iitialize variables t=. s_square_old =.! Sample fuctio do k=,trials xk=k! Sample Importace Fuctio g(x) usig rejectio call radom(rr)! Evaluate atithetic variates estimator! Atithetic variate! s = (f(rr)+f(.-rr))/.! Stratificatio s=(f(rr/.)+f(.5-rr/.)+f(.5+rr/.)+f(.-rr/.))/4. if (k.ge.) the ta = t/(xk-.) s_square_ew = (xk-.)/(xk-.)*s_square_old & & + (ta-s)*(ta-s)/xk s_square_old = s_square_ew ed if t=t+s ed do! Estimate mea value, variace, stadard deviatio,! ad fractioal stadard deviatio mea_value = t/trials std_dev = sqrt (s_square_ew) std_error = std_dev/sqrt(trials) frac_sta_dev = std_error/mea_value! Write results write(*,) mea_value, std_error, frac_sta_dev, trials write(44,) mea_value, std_error, frac_sta_dev, trials format(x,'mea value =',e4.8,/,x,'stadard error =',e4.8,/, & & x,'fractioal stadard deviatio =',e4.8,/, & & x,'umber of trials =',e4.8) ed Figure 9. Procedure for the applicatio of the Atithetic Variates ad Stratificatio Samplig techique with the t ad t estimators. Results are show i Tables 6 ad 7, ad demostrate the cosiderable advatage obtaied over other variace reductio methods. Table 6. Mea Values ad Stadard Deviatios for Atithetic Variates. Estimator t t f f / fsd , , Table 7. Mea Values ad Stadard Deviatios for Stratificatio.

17

18 For aother samplig scheme, the efficiecy is: (3) T These efficiecies are cosidered to be iversely proportioal to the product of the sample variace ad the amout of labor expeded i the computatio expressed i terms of the computatioal time T. This extra labor is ormally caused by the use of a more sophisticated samplig scheme. To compare two samplig methods ad, oe ca defie a figure of merit F or efficiecy ratio i terms of the ratio of their respective efficiecies as: T (4) VL T F It ca be oticed that the efficiecy ratio is the product of the variace ad labor ratios. The compariso i terms of the efficiecy ratio F ca be exteded to the compariso of the Mote Carlo method to aother umerical method if its error estimate ca be obtaied. A alterate defiitio of the efficiecy ratio ca be writte i terms of the stadard deviatios rather tha the variaces as: F ' T S L (5) T with a stadard deviatio ratio defied as: S (6). OVER-BIASING AND UNDER-BIASING: CHARACTERISTIC MONTE CARLO PITFALLS I modifyig the samplig for variace reductio, i.e. improved efficiecy or makig the best use of computig time to stimulate evets which are most sigificat to the fial aswer, it is possible to overshoot the mark ad produce a samplig scheme that is so strogly biased as to be less efficiet tha crude samplig. This is over-biasig or over-samplig. The opposite situatio is also ecoutered as uder-samplig or uder-biasig. This occurs i crude or a slightly modified samplig scheme whe the result depeds heavily o ifrequet evets ad ot eough observatios occurred for good statistics. It is a geeral characteristic of both cases that most of the time the geerated aswers are too small. This produces a apparetly cosistet bias i the result. Variace estimates are also geerally small so that the cofidece itervals calculated i the simulatio will ted to idicate that the results are much more accurate tha they

19 really are. This geerates a false feelig of security ad faith i the obtaied results, which are actually cosistetly bad. McGrath ad Irvig demostrate these ideas by a extremely simplified example: Cosider a simulatio with two types of evets. Oe type (X) occurs frequetly (f(x) =.999) but cotributes oly a small amout (g(x) =.) to the fial result, while the other type (X) is rare (f(x) =.) but makes a large cotributio (g(x) = ) whe it occurs. The quatity I beig estimated has the correct value: I =f(x )g(x ) + f(x )g(x ) =.9999x. +.x =. (7) Usig Crude Mote Carlo with a moderate umber of histories, evet X would rarely occur ad a uderbiased aswer would be recorded as: Iu=I g(x)=. (8) If importace samplig is used so that the X evets occurred frequetly; sice they cotribute so much to the aswer, (to a excess), say ew probabilities of f*(x) =.9999 ad f*(x) =., were used, the the X evets will ever occur i a ru of moderate size ad the overbiased result will be: I f( X). g X.. f ( X ).9999 (9) * The proper modificatio for that case is to let X ad X happe with equal probabilities, f * (X) = f * (X) =.5 (accordig to their cotributio to the fial aswer), the the cotributio from each history is: f( X ).9999 g X f X ( ).. * ( ).5 f( X ). g X ( ). * f ( X).5 (3) ad the fial estimate from a small sample would be: I =. The uderlyig distributio is highly skewed with the large majority of cases makig little or o cotributios to the fial aswer, while a small umber of cases ca make large cotributios. I both cases the fial estimates were smaller tha the correct aswer, which is a geeral characteristic of such cases. If a set of histories, say, was simulated by crude Mote Carlo, the most likely there would be o X evets observed ad the (icorrect) estimate will be.. Oce every sets of histories, a sigle evet X will happe ad the estimate will be:

20 ' 99. I u., (3) a umber very much larger tha the correct value. Eve though the estimatio procedure is right ad averages come out correctly i the log ru, most users would thik this was the result of some iput mistake or computatio error, ad throw out the ru. Therefore cautio is recommeded i simulatios where most histories cotribute a small bit to the aswer but a few histories cotribute a large value. Complete faith should ot be placed i estimates of variace especially whe the results are smaller tha expected or if the possibility of overbiasig is suspected. EXERCISES. Usig the Crude Mote Carlo method, write a procedure to estimate the value of the itegral: ( x) dx. Use a equatio for the estimatio of the variace of your choice. Plot the mea value ad the variace as a fuctio of the umber of experimets N.. Usig the correlated samplig variace reductio method, write a procedure to estimate the value of the itegral ( x) dx usig x e dx as a easy fuctio. Plot the mea value ad the variace as a fuctio of the umber of experimets N. Compare the result to that obtaied from the crude samplig Mote Carlo method. 3. Compare the results for the Correlated Samplig method usig the differet approximatio fuctios discussed. 4. Derive a higher level Atithetic Variates fuctio t3, ad compare its results to those obtaied usig the fuctios t ad t. 5. Write a procedure usig weighted Uiform Samplig for estimatig a itegral. Discuss ay produced bias i the result. 6. Compare the mea values, stadard errors, fractioal stadard deviatios, variace ratios, ad efficiecy ratios for the Crude Mote Carlo Samplig method for the estimatio of a itegral of your choice for differet umbers of trials. For istace you could use the itegral of x e dx over the iterval [,]. Now use the Atithetic Variates estimators t ad t, ad compare their results to each other ad to those obtaied usig the Crude Mote Carlo estimator. Choose the variace estimatio formula of your choice. Use the differece betwee two calls to the computer clock at the begiig ad ed of a calculatio to estimate the labor time.

Chapter 2 The Monte Carlo Method

Chapter 2 The Monte Carlo Method Chapter 2 The Mote Carlo Method The Mote Carlo Method stads for a broad class of computatioal algorithms that rely o radom sampligs. It is ofte used i physical ad mathematical problems ad is most useful

More information

Lecture 2: Monte Carlo Simulation

Lecture 2: Monte Carlo Simulation STAT/Q SCI 43: Itroductio to Resamplig ethods Sprig 27 Istructor: Ye-Chi Che Lecture 2: ote Carlo Simulatio 2 ote Carlo Itegratio Assume we wat to evaluate the followig itegratio: e x3 dx What ca we do?

More information

Clases 7-8: Métodos de reducción de varianza en Monte Carlo *

Clases 7-8: Métodos de reducción de varianza en Monte Carlo * Clases 7-8: Métodos de reducció de variaza e Mote Carlo * 9 de septiembre de 27 Ídice. Variace reductio 2. Atithetic variates 2 2.. Example: Uiform radom variables................ 3 2.2. Example: Tail

More information

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4

MATH 320: Probability and Statistics 9. Estimation and Testing of Parameters. Readings: Pruim, Chapter 4 MATH 30: Probability ad Statistics 9. Estimatio ad Testig of Parameters Estimatio ad Testig of Parameters We have bee dealig situatios i which we have full kowledge of the distributio of a radom variable.

More information

Simulation. Two Rule For Inverting A Distribution Function

Simulation. Two Rule For Inverting A Distribution Function Simulatio Two Rule For Ivertig A Distributio Fuctio Rule 1. If F(x) = u is costat o a iterval [x 1, x 2 ), the the uiform value u is mapped oto x 2 through the iversio process. Rule 2. If there is a jump

More information

Discrete Mathematics for CS Spring 2008 David Wagner Note 22

Discrete Mathematics for CS Spring 2008 David Wagner Note 22 CS 70 Discrete Mathematics for CS Sprig 2008 David Wager Note 22 I.I.D. Radom Variables Estimatig the bias of a coi Questio: We wat to estimate the proportio p of Democrats i the US populatio, by takig

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2016 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Random Variables, Sampling and Estimation

Random Variables, Sampling and Estimation Chapter 1 Radom Variables, Samplig ad Estimatio 1.1 Itroductio This chapter will cover the most importat basic statistical theory you eed i order to uderstad the ecoometric material that will be comig

More information

Monte Carlo method and application to random processes

Monte Carlo method and application to random processes Mote Carlo method ad applicatio to radom processes Lecture 3: Variace reductio techiques (8/3/2017) 1 Lecturer: Eresto Mordecki, Facultad de Ciecias, Uiversidad de la República, Motevideo, Uruguay Graduate

More information

DS 100: Principles and Techniques of Data Science Date: April 13, Discussion #10

DS 100: Principles and Techniques of Data Science Date: April 13, Discussion #10 DS 00: Priciples ad Techiques of Data Sciece Date: April 3, 208 Name: Hypothesis Testig Discussio #0. Defie these terms below as they relate to hypothesis testig. a) Data Geeratio Model: Solutio: A set

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Monte Carlo Integration

Monte Carlo Integration Mote Carlo Itegratio I these otes we first review basic umerical itegratio methods (usig Riema approximatio ad the trapezoidal rule) ad their limitatios for evaluatig multidimesioal itegrals. Next we itroduce

More information

1 Introduction to reducing variance in Monte Carlo simulations

1 Introduction to reducing variance in Monte Carlo simulations Copyright c 010 by Karl Sigma 1 Itroductio to reducig variace i Mote Carlo simulatios 11 Review of cofidece itervals for estimatig a mea I statistics, we estimate a ukow mea µ = E(X) of a distributio by

More information

Resampling Methods. X (1/2), i.e., Pr (X i m) = 1/2. We order the data: X (1) X (2) X (n). Define the sample median: ( n.

Resampling Methods. X (1/2), i.e., Pr (X i m) = 1/2. We order the data: X (1) X (2) X (n). Define the sample median: ( n. Jauary 1, 2019 Resamplig Methods Motivatio We have so may estimators with the property θ θ d N 0, σ 2 We ca also write θ a N θ, σ 2 /, where a meas approximately distributed as Oce we have a cosistet estimator

More information

4. Partial Sums and the Central Limit Theorem

4. Partial Sums and the Central Limit Theorem 1 of 10 7/16/2009 6:05 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 4. Partial Sums ad the Cetral Limit Theorem The cetral limit theorem ad the law of large umbers are the two fudametal theorems

More information

Computing Confidence Intervals for Sample Data

Computing Confidence Intervals for Sample Data Computig Cofidece Itervals for Sample Data Topics Use of Statistics Sources of errors Accuracy, precisio, resolutio A mathematical model of errors Cofidece itervals For meas For variaces For proportios

More information

Hypothesis Testing. Evaluation of Performance of Learned h. Issues. Trade-off Between Bias and Variance

Hypothesis Testing. Evaluation of Performance of Learned h. Issues. Trade-off Between Bias and Variance Hypothesis Testig Empirically evaluatig accuracy of hypotheses: importat activity i ML. Three questios: Give observed accuracy over a sample set, how well does this estimate apply over additioal samples?

More information

Topic 9: Sampling Distributions of Estimators

Topic 9: Sampling Distributions of Estimators Topic 9: Samplig Distributios of Estimators Course 003, 2018 Page 0 Samplig distributios of estimators Sice our estimators are statistics (particular fuctios of radom variables), their distributio ca be

More information

Machine Learning Brett Bernstein

Machine Learning Brett Bernstein Machie Learig Brett Berstei Week 2 Lecture: Cocept Check Exercises Starred problems are optioal. Excess Risk Decompositio 1. Let X = Y = {1, 2,..., 10}, A = {1,..., 10, 11} ad suppose the data distributio

More information

Since X n /n P p, we know that X n (n. Xn (n X n ) Using the asymptotic result above to obtain an approximation for fixed n, we obtain

Since X n /n P p, we know that X n (n. Xn (n X n ) Using the asymptotic result above to obtain an approximation for fixed n, we obtain Assigmet 9 Exercise 5.5 Let X biomial, p, where p 0, 1 is ukow. Obtai cofidece itervals for p i two differet ways: a Sice X / p d N0, p1 p], the variace of the limitig distributio depeds oly o p. Use the

More information

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample.

Statistical Inference (Chapter 10) Statistical inference = learn about a population based on the information provided by a sample. Statistical Iferece (Chapter 10) Statistical iferece = lear about a populatio based o the iformatio provided by a sample. Populatio: The set of all values of a radom variable X of iterest. Characterized

More information

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures

FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING. Lectures FACULTY OF MATHEMATICAL STUDIES MATHEMATICS FOR PART I ENGINEERING Lectures MODULE 5 STATISTICS II. Mea ad stadard error of sample data. Biomial distributio. Normal distributio 4. Samplig 5. Cofidece itervals

More information

This exam contains 19 pages (including this cover page) and 10 questions. A Formulae sheet is provided with the exam.

This exam contains 19 pages (including this cover page) and 10 questions. A Formulae sheet is provided with the exam. Probability ad Statistics FS 07 Secod Sessio Exam 09.0.08 Time Limit: 80 Miutes Name: Studet ID: This exam cotais 9 pages (icludig this cover page) ad 0 questios. A Formulae sheet is provided with the

More information

1 of 7 7/16/2009 6:06 AM Virtual Laboratories > 6. Radom Samples > 1 2 3 4 5 6 7 6. Order Statistics Defiitios Suppose agai that we have a basic radom experimet, ad that X is a real-valued radom variable

More information

Output Analysis (2, Chapters 10 &11 Law)

Output Analysis (2, Chapters 10 &11 Law) B. Maddah ENMG 6 Simulatio Output Aalysis (, Chapters 10 &11 Law) Comparig alterative system cofiguratio Sice the output of a simulatio is radom, the comparig differet systems via simulatio should be doe

More information

Expectation and Variance of a random variable

Expectation and Variance of a random variable Chapter 11 Expectatio ad Variace of a radom variable The aim of this lecture is to defie ad itroduce mathematical Expectatio ad variace of a fuctio of discrete & cotiuous radom variables ad the distributio

More information

ENGI 4421 Confidence Intervals (Two Samples) Page 12-01

ENGI 4421 Confidence Intervals (Two Samples) Page 12-01 ENGI 44 Cofidece Itervals (Two Samples) Page -0 Two Sample Cofidece Iterval for a Differece i Populatio Meas [Navidi sectios 5.4-5.7; Devore chapter 9] From the cetral limit theorem, we kow that, for sufficietly

More information

Lecture 19: Convergence

Lecture 19: Convergence Lecture 19: Covergece Asymptotic approach I statistical aalysis or iferece, a key to the success of fidig a good procedure is beig able to fid some momets ad/or distributios of various statistics. I may

More information

Topic 10: The Law of Large Numbers

Topic 10: The Law of Large Numbers Topic : October 6, 2 If we choose adult Europea males idepedetly ad measure their heights, keepig a ruig average, the at the begiig we might see some larger fluctuatios but as we cotiue to make measuremets,

More information

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara

Econ 325 Notes on Point Estimator and Confidence Interval 1 By Hiro Kasahara Poit Estimator Eco 325 Notes o Poit Estimator ad Cofidece Iterval 1 By Hiro Kasahara Parameter, Estimator, ad Estimate The ormal probability desity fuctio is fully characterized by two costats: populatio

More information

n outcome is (+1,+1, 1,..., 1). Let the r.v. X denote our position (relative to our starting point 0) after n moves. Thus X = X 1 + X 2 + +X n,

n outcome is (+1,+1, 1,..., 1). Let the r.v. X denote our position (relative to our starting point 0) after n moves. Thus X = X 1 + X 2 + +X n, CS 70 Discrete Mathematics for CS Sprig 2008 David Wager Note 9 Variace Questio: At each time step, I flip a fair coi. If it comes up Heads, I walk oe step to the right; if it comes up Tails, I walk oe

More information

Lecture 7: Properties of Random Samples

Lecture 7: Properties of Random Samples Lecture 7: Properties of Radom Samples 1 Cotiued From Last Class Theorem 1.1. Let X 1, X,...X be a radom sample from a populatio with mea µ ad variace σ

More information

Sample Size Determination (Two or More Samples)

Sample Size Determination (Two or More Samples) Sample Sie Determiatio (Two or More Samples) STATGRAPHICS Rev. 963 Summary... Data Iput... Aalysis Summary... 5 Power Curve... 5 Calculatios... 6 Summary This procedure determies a suitable sample sie

More information

Stat 421-SP2012 Interval Estimation Section

Stat 421-SP2012 Interval Estimation Section Stat 41-SP01 Iterval Estimatio Sectio 11.1-11. We ow uderstad (Chapter 10) how to fid poit estimators of a ukow parameter. o However, a poit estimate does ot provide ay iformatio about the ucertaity (possible

More information

Distribution of Random Samples & Limit theorems

Distribution of Random Samples & Limit theorems STAT/MATH 395 A - PROBABILITY II UW Witer Quarter 2017 Néhémy Lim Distributio of Radom Samples & Limit theorems 1 Distributio of i.i.d. Samples Motivatig example. Assume that the goal of a study is to

More information

Unbiased Estimation. February 7-12, 2008

Unbiased Estimation. February 7-12, 2008 Ubiased Estimatio February 7-2, 2008 We begi with a sample X = (X,..., X ) of radom variables chose accordig to oe of a family of probabilities P θ where θ is elemet from the parameter space Θ. For radom

More information

Comparing your lab results with the others by one-way ANOVA

Comparing your lab results with the others by one-way ANOVA Comparig your lab results with the others by oe-way ANOVA You may have developed a ew test method ad i your method validatio process you would like to check the method s ruggedess by coductig a simple

More information

BIOSTATISTICS. Lecture 5 Interval Estimations for Mean and Proportion. dr. Petr Nazarov

BIOSTATISTICS. Lecture 5 Interval Estimations for Mean and Proportion. dr. Petr Nazarov Microarray Ceter BIOSTATISTICS Lecture 5 Iterval Estimatios for Mea ad Proportio dr. Petr Nazarov 15-03-013 petr.azarov@crp-sate.lu Lecture 5. Iterval estimatio for mea ad proportio OUTLINE Iterval estimatios

More information

Tests of Hypotheses Based on a Single Sample (Devore Chapter Eight)

Tests of Hypotheses Based on a Single Sample (Devore Chapter Eight) Tests of Hypotheses Based o a Sigle Sample Devore Chapter Eight MATH-252-01: Probability ad Statistics II Sprig 2018 Cotets 1 Hypothesis Tests illustrated with z-tests 1 1.1 Overview of Hypothesis Testig..........

More information

(7 One- and Two-Sample Estimation Problem )

(7 One- and Two-Sample Estimation Problem ) 34 Stat Lecture Notes (7 Oe- ad Two-Sample Estimatio Problem ) ( Book*: Chapter 8,pg65) Probability& Statistics for Egieers & Scietists By Walpole, Myers, Myers, Ye Estimatio 1 ) ( ˆ S P i i Poit estimate:

More information

April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE

April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE April 18, 2017 CONFIDENCE INTERVALS AND HYPOTHESIS TESTING, UNDERGRADUATE MATH 526 STYLE TERRY SOO Abstract These otes are adapted from whe I taught Math 526 ad meat to give a quick itroductio to cofidece

More information

Stochastic Simulation

Stochastic Simulation Stochastic Simulatio 1 Itroductio Readig Assigmet: Read Chapter 1 of text. We shall itroduce may of the key issues to be discussed i this course via a couple of model problems. Model Problem 1 (Jackso

More information

MATH/STAT 352: Lecture 15

MATH/STAT 352: Lecture 15 MATH/STAT 352: Lecture 15 Sectios 5.2 ad 5.3. Large sample CI for a proportio ad small sample CI for a mea. 1 5.2: Cofidece Iterval for a Proportio Estimatig proportio of successes i a biomial experimet

More information

Review Questions, Chapters 8, 9. f(y) = 0, elsewhere. F (y) = f Y(1) = n ( e y/θ) n 1 1 θ e y/θ = n θ e yn

Review Questions, Chapters 8, 9. f(y) = 0, elsewhere. F (y) = f Y(1) = n ( e y/θ) n 1 1 θ e y/θ = n θ e yn Stat 366 Lab 2 Solutios (September 2, 2006) page TA: Yury Petracheko, CAB 484, yuryp@ualberta.ca, http://www.ualberta.ca/ yuryp/ Review Questios, Chapters 8, 9 8.5 Suppose that Y, Y 2,..., Y deote a radom

More information

NANYANG TECHNOLOGICAL UNIVERSITY SYLLABUS FOR ENTRANCE EXAMINATION FOR INTERNATIONAL STUDENTS AO-LEVEL MATHEMATICS

NANYANG TECHNOLOGICAL UNIVERSITY SYLLABUS FOR ENTRANCE EXAMINATION FOR INTERNATIONAL STUDENTS AO-LEVEL MATHEMATICS NANYANG TECHNOLOGICAL UNIVERSITY SYLLABUS FOR ENTRANCE EXAMINATION FOR INTERNATIONAL STUDENTS AO-LEVEL MATHEMATICS STRUCTURE OF EXAMINATION PAPER. There will be oe 2-hour paper cosistig of 4 questios.

More information

Using the IML Procedure to Examine the Efficacy of a New Control Charting Technique

Using the IML Procedure to Examine the Efficacy of a New Control Charting Technique Paper 2894-2018 Usig the IML Procedure to Examie the Efficacy of a New Cotrol Chartig Techique Austi Brow, M.S., Uiversity of Norther Colorado; Bryce Whitehead, M.S., Uiversity of Norther Colorado ABSTRACT

More information

Understanding Samples

Understanding Samples 1 Will Moroe CS 109 Samplig ad Bootstrappig Lecture Notes #17 August 2, 2017 Based o a hadout by Chris Piech I this chapter we are goig to talk about statistics calculated o samples from a populatio. We

More information

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f.

Let us give one more example of MLE. Example 3. The uniform distribution U[0, θ] on the interval [0, θ] has p.d.f. Lecture 5 Let us give oe more example of MLE. Example 3. The uiform distributio U[0, ] o the iterval [0, ] has p.d.f. { 1 f(x =, 0 x, 0, otherwise The likelihood fuctio ϕ( = f(x i = 1 I(X 1,..., X [0,

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit Theorems Throughout this sectio we will assume a probability space (, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

ANALYSIS OF EXPERIMENTAL ERRORS

ANALYSIS OF EXPERIMENTAL ERRORS ANALYSIS OF EXPERIMENTAL ERRORS All physical measuremets ecoutered i the verificatio of physics theories ad cocepts are subject to ucertaities that deped o the measurig istrumets used ad the coditios uder

More information

Last time: Moments of the Poisson distribution from its generating function. Example: Using telescope to measure intensity of an object

Last time: Moments of the Poisson distribution from its generating function. Example: Using telescope to measure intensity of an object 6.3 Stochastic Estimatio ad Cotrol, Fall 004 Lecture 7 Last time: Momets of the Poisso distributio from its geeratig fuctio. Gs () e dg µ e ds dg µ ( s) µ ( s) µ ( s) µ e ds dg X µ ds X s dg dg + ds ds

More information

6.3 Testing Series With Positive Terms

6.3 Testing Series With Positive Terms 6.3. TESTING SERIES WITH POSITIVE TERMS 307 6.3 Testig Series With Positive Terms 6.3. Review of what is kow up to ow I theory, testig a series a i for covergece amouts to fidig the i= sequece of partial

More information

7-1. Chapter 4. Part I. Sampling Distributions and Confidence Intervals

7-1. Chapter 4. Part I. Sampling Distributions and Confidence Intervals 7-1 Chapter 4 Part I. Samplig Distributios ad Cofidece Itervals 1 7- Sectio 1. Samplig Distributio 7-3 Usig Statistics Statistical Iferece: Predict ad forecast values of populatio parameters... Test hypotheses

More information

Module 1 Fundamentals in statistics

Module 1 Fundamentals in statistics Normal Distributio Repeated observatios that differ because of experimetal error ofte vary about some cetral value i a roughly symmetrical distributio i which small deviatios occur much more frequetly

More information

CS / MCS 401 Homework 3 grader solutions

CS / MCS 401 Homework 3 grader solutions CS / MCS 401 Homework 3 grader solutios assigmet due July 6, 016 writte by Jāis Lazovskis maximum poits: 33 Some questios from CLRS. Questios marked with a asterisk were ot graded. 1 Use the defiitio of

More information

STA Learning Objectives. Population Proportions. Module 10 Comparing Two Proportions. Upon completing this module, you should be able to:

STA Learning Objectives. Population Proportions. Module 10 Comparing Two Proportions. Upon completing this module, you should be able to: STA 2023 Module 10 Comparig Two Proportios Learig Objectives Upo completig this module, you should be able to: 1. Perform large-sample ifereces (hypothesis test ad cofidece itervals) to compare two populatio

More information

Chapter 9: Numerical Differentiation

Chapter 9: Numerical Differentiation 178 Chapter 9: Numerical Differetiatio Numerical Differetiatio Formulatio of equatios for physical problems ofte ivolve derivatives (rate-of-chage quatities, such as velocity ad acceleratio). Numerical

More information

Lecture 33: Bootstrap

Lecture 33: Bootstrap Lecture 33: ootstrap Motivatio To evaluate ad compare differet estimators, we eed cosistet estimators of variaces or asymptotic variaces of estimators. This is also importat for hypothesis testig ad cofidece

More information

Exponential Families and Bayesian Inference

Exponential Families and Bayesian Inference Computer Visio Expoetial Families ad Bayesia Iferece Lecture Expoetial Families A expoetial family of distributios is a d-parameter family f(x; havig the followig form: f(x; = h(xe g(t T (x B(, (. where

More information

Chapter 6 Sampling Distributions

Chapter 6 Sampling Distributions Chapter 6 Samplig Distributios 1 I most experimets, we have more tha oe measuremet for ay give variable, each measuremet beig associated with oe radomly selected a member of a populatio. Hece we eed to

More information

If, for instance, we were required to test whether the population mean μ could be equal to a certain value μ

If, for instance, we were required to test whether the population mean μ could be equal to a certain value μ STATISTICAL INFERENCE INTRODUCTION Statistical iferece is that brach of Statistics i which oe typically makes a statemet about a populatio based upo the results of a sample. I oesample testig, we essetially

More information

Confidence intervals summary Conservative and approximate confidence intervals for a binomial p Examples. MATH1005 Statistics. Lecture 24. M.

Confidence intervals summary Conservative and approximate confidence intervals for a binomial p Examples. MATH1005 Statistics. Lecture 24. M. MATH1005 Statistics Lecture 24 M. Stewart School of Mathematics ad Statistics Uiversity of Sydey Outlie Cofidece itervals summary Coservative ad approximate cofidece itervals for a biomial p The aïve iterval

More information

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting Lecture 6 Chi Square Distributio (χ ) ad Least Squares Fittig Chi Square Distributio (χ ) Suppose: We have a set of measuremets {x 1, x, x }. We kow the true value of each x i (x t1, x t, x t ). We would

More information

Estimation of a population proportion March 23,

Estimation of a population proportion March 23, 1 Social Studies 201 Notes for March 23, 2005 Estimatio of a populatio proportio Sectio 8.5, p. 521. For the most part, we have dealt with meas ad stadard deviatios this semester. This sectio of the otes

More information

Estimation for Complete Data

Estimation for Complete Data Estimatio for Complete Data complete data: there is o loss of iformatio durig study. complete idividual complete data= grouped data A complete idividual data is the oe i which the complete iformatio of

More information

Statisticians use the word population to refer the total number of (potential) observations under consideration

Statisticians use the word population to refer the total number of (potential) observations under consideration 6 Samplig Distributios Statisticias use the word populatio to refer the total umber of (potetial) observatios uder cosideratio The populatio is just the set of all possible outcomes i our sample space

More information

Topic 10: Introduction to Estimation

Topic 10: Introduction to Estimation Topic 0: Itroductio to Estimatio Jue, 0 Itroductio I the simplest possible terms, the goal of estimatio theory is to aswer the questio: What is that umber? What is the legth, the reactio rate, the fractio

More information

MBACATÓLICA. Quantitative Methods. Faculdade de Ciências Económicas e Empresariais UNIVERSIDADE CATÓLICA PORTUGUESA 9. SAMPLING DISTRIBUTIONS

MBACATÓLICA. Quantitative Methods. Faculdade de Ciências Económicas e Empresariais UNIVERSIDADE CATÓLICA PORTUGUESA 9. SAMPLING DISTRIBUTIONS MBACATÓLICA Quatitative Methods Miguel Gouveia Mauel Leite Moteiro Faculdade de Ciêcias Ecoómicas e Empresariais UNIVERSIDADE CATÓLICA PORTUGUESA 9. SAMPLING DISTRIBUTIONS MBACatólica 006/07 Métodos Quatitativos

More information

Sample Size Estimation in the Proportional Hazards Model for K-sample or Regression Settings Scott S. Emerson, M.D., Ph.D.

Sample Size Estimation in the Proportional Hazards Model for K-sample or Regression Settings Scott S. Emerson, M.D., Ph.D. ample ie Estimatio i the Proportioal Haards Model for K-sample or Regressio ettigs cott. Emerso, M.D., Ph.D. ample ie Formula for a Normally Distributed tatistic uppose a statistic is kow to be ormally

More information

Chapter 6 Principles of Data Reduction

Chapter 6 Principles of Data Reduction Chapter 6 for BST 695: Special Topics i Statistical Theory. Kui Zhag, 0 Chapter 6 Priciples of Data Reductio Sectio 6. Itroductio Goal: To summarize or reduce the data X, X,, X to get iformatio about a

More information

Surveying the Variance Reduction Methods

Surveying the Variance Reduction Methods Iteratioal Research Joural of Applied ad Basic Scieces 2013 Available olie at www.irjabs.com ISSN 2251-838X / Vol, 7 (7): 427-432 Sciece Explorer Publicatios Surveyig the Variace Reductio Methods Arash

More information

Department of Mathematics

Department of Mathematics Departmet of Mathematics Ma 3/103 KC Border Itroductio to Probability ad Statistics Witer 2017 Lecture 19: Estimatio II Relevat textbook passages: Larse Marx [1]: Sectios 5.2 5.7 19.1 The method of momets

More information

Frequentist Inference

Frequentist Inference Frequetist Iferece The topics of the ext three sectios are useful applicatios of the Cetral Limit Theorem. Without kowig aythig about the uderlyig distributio of a sequece of radom variables {X i }, for

More information

1 Inferential Methods for Correlation and Regression Analysis

1 Inferential Methods for Correlation and Regression Analysis 1 Iferetial Methods for Correlatio ad Regressio Aalysis I the chapter o Correlatio ad Regressio Aalysis tools for describig bivariate cotiuous data were itroduced. The sample Pearso Correlatio Coefficiet

More information

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting

Lecture 6 Chi Square Distribution (χ 2 ) and Least Squares Fitting Lecture 6 Chi Square Distributio (χ ) ad Least Squares Fittig Chi Square Distributio (χ ) Suppose: We have a set of measuremets {x 1, x, x }. We kow the true value of each x i (x t1, x t, x t ). We would

More information

Sampling Error. Chapter 6 Student Lecture Notes 6-1. Business Statistics: A Decision-Making Approach, 6e. Chapter Goals

Sampling Error. Chapter 6 Student Lecture Notes 6-1. Business Statistics: A Decision-Making Approach, 6e. Chapter Goals Chapter 6 Studet Lecture Notes 6-1 Busiess Statistics: A Decisio-Makig Approach 6 th Editio Chapter 6 Itroductio to Samplig Distributios Chap 6-1 Chapter Goals After completig this chapter, you should

More information

Poisson approximations

Poisson approximations The Bi, p) ca be thought of as the distributio of a sum of idepedet idicator radom variables X +...+ X, with {X i = } deotig a head o the ith toss of a coi. The ormal approximatio to the Biomial works

More information

PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 9

PSYCHOLOGICAL RESEARCH (PYC 304-C) Lecture 9 Hypothesis testig PSYCHOLOGICAL RESEARCH (PYC 34-C Lecture 9 Statistical iferece is that brach of Statistics i which oe typically makes a statemet about a populatio based upo the results of a sample. I

More information

Statistical inference: example 1. Inferential Statistics

Statistical inference: example 1. Inferential Statistics Statistical iferece: example 1 Iferetial Statistics POPULATION SAMPLE A clothig store chai regularly buys from a supplier large quatities of a certai piece of clothig. Each item ca be classified either

More information

Binomial Distribution

Binomial Distribution 0.0 0.5 1.0 1.5 2.0 2.5 3.0 0 1 2 3 4 5 6 7 0.0 0.5 1.0 1.5 2.0 2.5 3.0 Overview Example: coi tossed three times Defiitio Formula Recall that a r.v. is discrete if there are either a fiite umber of possible

More information

Problem Set 4 Due Oct, 12

Problem Set 4 Due Oct, 12 EE226: Radom Processes i Systems Lecturer: Jea C. Walrad Problem Set 4 Due Oct, 12 Fall 06 GSI: Assae Gueye This problem set essetially reviews detectio theory ad hypothesis testig ad some basic otios

More information

IE 230 Probability & Statistics in Engineering I. Closed book and notes. No calculators. 120 minutes.

IE 230 Probability & Statistics in Engineering I. Closed book and notes. No calculators. 120 minutes. Closed book ad otes. No calculators. 120 miutes. Cover page, five pages of exam, ad tables for discrete ad cotiuous distributios. Score X i =1 X i / S X 2 i =1 (X i X ) 2 / ( 1) = [i =1 X i 2 X 2 ] / (

More information

Properties and Hypothesis Testing

Properties and Hypothesis Testing Chapter 3 Properties ad Hypothesis Testig 3.1 Types of data The regressio techiques developed i previous chapters ca be applied to three differet kids of data. 1. Cross-sectioal data. 2. Time series data.

More information

Sampling Distributions, Z-Tests, Power

Sampling Distributions, Z-Tests, Power Samplig Distributios, Z-Tests, Power We draw ifereces about populatio parameters from sample statistics Sample proportio approximates populatio proportio Sample mea approximates populatio mea Sample variace

More information

6. Sufficient, Complete, and Ancillary Statistics

6. Sufficient, Complete, and Ancillary Statistics Sufficiet, Complete ad Acillary Statistics http://www.math.uah.edu/stat/poit/sufficiet.xhtml 1 of 7 7/16/2009 6:13 AM Virtual Laboratories > 7. Poit Estimatio > 1 2 3 4 5 6 6. Sufficiet, Complete, ad Acillary

More information

1 Approximating Integrals using Taylor Polynomials

1 Approximating Integrals using Taylor Polynomials Seughee Ye Ma 8: Week 7 Nov Week 7 Summary This week, we will lear how we ca approximate itegrals usig Taylor series ad umerical methods. Topics Page Approximatig Itegrals usig Taylor Polyomials. Defiitios................................................

More information

KLMED8004 Medical statistics. Part I, autumn Estimation. We have previously learned: Population and sample. New questions

KLMED8004 Medical statistics. Part I, autumn Estimation. We have previously learned: Population and sample. New questions We have previously leared: KLMED8004 Medical statistics Part I, autum 00 How kow probability distributios (e.g. biomial distributio, ormal distributio) with kow populatio parameters (mea, variace) ca give

More information

The standard deviation of the mean

The standard deviation of the mean Physics 6C Fall 20 The stadard deviatio of the mea These otes provide some clarificatio o the distictio betwee the stadard deviatio ad the stadard deviatio of the mea.. The sample mea ad variace Cosider

More information

7.1 Convergence of sequences of random variables

7.1 Convergence of sequences of random variables Chapter 7 Limit theorems Throughout this sectio we will assume a probability space (Ω, F, P), i which is defied a ifiite sequece of radom variables (X ) ad a radom variable X. The fact that for every ifiite

More information

Basics of Probability Theory (for Theory of Computation courses)

Basics of Probability Theory (for Theory of Computation courses) Basics of Probability Theory (for Theory of Computatio courses) Oded Goldreich Departmet of Computer Sciece Weizma Istitute of Sciece Rehovot, Israel. oded.goldreich@weizma.ac.il November 24, 2008 Preface.

More information

An Introduction to Randomized Algorithms

An Introduction to Randomized Algorithms A Itroductio to Radomized Algorithms The focus of this lecture is to study a radomized algorithm for quick sort, aalyze it usig probabilistic recurrece relatios, ad also provide more geeral tools for aalysis

More information

LECTURE 8: ASYMPTOTICS I

LECTURE 8: ASYMPTOTICS I LECTURE 8: ASYMPTOTICS I We are iterested i the properties of estimators as. Cosider a sequece of radom variables {, X 1}. N. M. Kiefer, Corell Uiversity, Ecoomics 60 1 Defiitio: (Weak covergece) A sequece

More information

Math 10A final exam, December 16, 2016

Math 10A final exam, December 16, 2016 Please put away all books, calculators, cell phoes ad other devices. You may cosult a sigle two-sided sheet of otes. Please write carefully ad clearly, USING WORDS (ot just symbols). Remember that the

More information

BUSINESS STATISTICS (PART-9) AVERAGE OR MEASURES OF CENTRAL TENDENCY: THE GEOMETRIC AND HARMONIC MEANS

BUSINESS STATISTICS (PART-9) AVERAGE OR MEASURES OF CENTRAL TENDENCY: THE GEOMETRIC AND HARMONIC MEANS BUSINESS STATISTICS (PART-9) AVERAGE OR MEASURES OF CENTRAL TENDENCY: THE GEOMETRIC AND HARMONIC MEANS. INTRODUCTION We have so far discussed three measures of cetral tedecy, viz. The Arithmetic Mea, Media

More information

IE 230 Seat # Name < KEY > Please read these directions. Closed book and notes. 60 minutes.

IE 230 Seat # Name < KEY > Please read these directions. Closed book and notes. 60 minutes. IE 230 Seat # Name < KEY > Please read these directios. Closed book ad otes. 60 miutes. Covers through the ormal distributio, Sectio 4.7 of Motgomery ad Ruger, fourth editio. Cover page ad four pages of

More information

A statistical method to determine sample size to estimate characteristic value of soil parameters

A statistical method to determine sample size to estimate characteristic value of soil parameters A statistical method to determie sample size to estimate characteristic value of soil parameters Y. Hojo, B. Setiawa 2 ad M. Suzuki 3 Abstract Sample size is a importat factor to be cosidered i determiig

More information

This is an introductory course in Analysis of Variance and Design of Experiments.

This is an introductory course in Analysis of Variance and Design of Experiments. 1 Notes for M 384E, Wedesday, Jauary 21, 2009 (Please ote: I will ot pass out hard-copy class otes i future classes. If there are writte class otes, they will be posted o the web by the ight before class

More information

PRACTICE PROBLEMS FOR THE FINAL

PRACTICE PROBLEMS FOR THE FINAL PRACTICE PROBLEMS FOR THE FINAL Math 36Q Fall 25 Professor Hoh Below is a list of practice questios for the Fial Exam. I would suggest also goig over the practice problems ad exams for Exam ad Exam 2 to

More information

A Question. Output Analysis. Example. What Are We Doing Wrong? Result from throwing a die. Let X be the random variable

A Question. Output Analysis. Example. What Are We Doing Wrong? Result from throwing a die. Let X be the random variable A Questio Output Aalysis Let X be the radom variable Result from throwig a die 5.. Questio: What is E (X? Would you throw just oce ad take the result as your aswer? Itroductio to Simulatio WS/ - L 7 /

More information

Analysis of Experimental Measurements

Analysis of Experimental Measurements Aalysis of Experimetal Measuremets Thik carefully about the process of makig a measuremet. A measuremet is a compariso betwee some ukow physical quatity ad a stadard of that physical quatity. As a example,

More information