47 Pages

Week3

Course: FOMGT 421, Fall 2009
School: UMass (Amherst)
Rating:
 
 
 
 
 

Word Count: 3134

Document Preview

to Introduction Mutual Funds Basic Portfolio Mathematics Week 3: 1 An Example of A Mutual Fund The largest mutual fund is the Fidelity Magellan Fund, with assets of $76.885 billion (31/1/2002). The fund has been in existence since May 1963. It is currently closed to most new investment. What type of a fund is it? It invests in large caps, and blend of growth and value. Given its style, what should its...

Register Now

Unformatted Document Excerpt

Coursehero >> Massachusetts >> UMass (Amherst) >> FOMGT 421

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
to Introduction Mutual Funds Basic Portfolio Mathematics Week 3: 1 An Example of A Mutual Fund The largest mutual fund is the Fidelity Magellan Fund, with assets of $76.885 billion (31/1/2002). The fund has been in existence since May 1963. It is currently closed to most new investment. What type of a fund is it? It invests in large caps, and blend of growth and value. Given its style, what should its benchmark be? The appropriate benchmark, because of its emphasis on large caps, is the S&P 500. What kind of stocks would you buy if you were the manager of Magellan? 2 Magellan's Stock Holding on 12/31/01 1.GENERAL ELECTRIC CO (4.76%) 2. CITIGROUP INC (3.95%) 3. MICROSOFT CORP 4. TYCO INTL LTD (2.69%) 5. AMER INTL GROUP INC (3%) 6. VIACOM INC CL B NON-VTG 7. EXXON MOBIL CORP (2.83%) 8. PFIZER INC 9. WAL MART STORES INC 10. HOME DEPOT INC 3 Magellan vs. the S&P 500 29.50% of Magellan's holdings were in these top 10 stocks on 12/31/01. First, note that many of the stocks in the top holdings match the stocks with the highest weight in the S&P 500. Which stocks are missing from Fidelity's holdings Intel and IBM so it appears that Fidelity is underweighted in technology. Second, the weights are different. The S&P 500 had a weight of 22.09% in these 10 stocks. (Given these weights of Magellan, what do you estimate Magellan's performance, year to date, compared to its benchmark?). 4 Magellan vs. S&P 500 vs. Average Fund in Group (as of 31 Dec 2001) Last 1 year: Magellan = -11.65%. S&P 500 = -11.89%. Average Growth Fund = -16.76%. Last 5 years. Magellan = 10.95%. S&P 500 = 10.70%. Average Growth Fund = 8.56%. Last 10 years. Magellan = 12.90%. S&P 500 = 12.94%. Average Growth Fund = 11.19%. But Fidelity Magellan charges a "front-end load" a fee of 3% for entering the fund. The 1, 5, 10 year returns after the load are: -14.30%, 10.28%, 12.56%. So, after accounting for the load, Magellan underperforms the S&P 500 over each of these periods http://personal300.fidelity.com/products/funds/mfl_frame.shtml? 316184100. 5 What Magellan Charges for Managing Your Money According to its annual report, 3/31/2001: Management fee Basic fee = $ 571,113,000. Performance adjustment = $139,203,000. Besides the management fee, the fund will charge other operational expenses. Total expenses, including management fee, added up to: 872,538,000. The ratio of expenses to net assets = 0.89%. If Magellan was open to new investment, it could have charged an additional fee, if necessary, called the 12B-1. This fee could be used for marketing purposes. Currently, Magellan has no 12B-1 fees. Finally, Magellan can charge a "load" either a front-end or a back-end load. Magellan has a 3% front-end load. 6 Fidelity Magellan Year Ending 31/3/2001 NAV (Beginning of Year) Investment Operations Net Investment Income (dividends,etc.) Net Realized and Unrealized Gain on Investment Distributions Dividends Distributions from Capital Gains NAV( End of Year) (Net Assets of $80,190,261,000/ 767,360 shares) 2001 143.26 0.37 -34.17 -0.27 -4.69 $104.50 Ratios/Supplemental Data Ratio of Total Expenses to Average Net Asset Ratio of Net Investment Income to Average Net Assets Portfolio Turnover Rate 0.89% 0.59% 24% 7 Understanding the Numbers (1/2) New NAV: Old NAV + investment income + net realized and unrealized gain - all distributions. Distributions: To avoid taxation at the fund level, the fund must pass on any dividend or capital gains directly to the investors. The investors will now pay tax at their personal rate on both the dividends and capital gains. Fidelity has distributed $4.96 per share. Expense Ratio: This summarizes the operating expenses of the fund as a fraction of its NAV. The Magellan Fund has an expense ratio of 0.89%. This is comparable with other funds, but appear high relative to its size. As we saw, this is equal to $872 million. 8 Understanding the Numbers Portfolio Turnover Rate: This represents the fraction of the portfolio that is sold during the year. A turnover rate of 24% indicates that the average stock was held for 1/0.24=4.16 years. To see the effect of other potential charges, in particular loads and 12b-1 fees, let us consider another example. 9 The Types of Fees Charged by Funds: Loads and Fees Loads: front end or back end Fees: Management Fee 12B-1 Fees Other expenses Consider, as an example, the Oppenheimer Funds. 10 Fund Fees: Loads (1/2): Oppenheimer Growth Fund. Front End Load: A commission or sales charge paid when the shares of the fund are purchased. For example, Oppenheimer Funds have a typical front end load of 5.75% for their Class A shares. Back End Load: This is a redemption or exit fee that is paid when the funds are withdrawn. For example, Oppenheimer charges a 5% back end fee for its Class B shares, that decrease to 1% and is eliminated from 6th year onwards. Oppenheimer's Class B shares are converted automatically to Class A shares at the end of the 6th year. 11 Fund Fees: Operating Expenses(2/2) Annual Fund Operating Expenses: Management Fee + 12b-1 + Other operating expenses. 12b-1 Charges: The fund may charge a 12b-1 fee for marketing and advertising expenses, as well as commissions paid to brokers that sell the fund. This can be in addition to a front-end/back-end load. Oppenheimer charges a 12b-1 fee of 1% for both its Class B and C shares, and a fee of 0.25% for its Class A shares. Management Fee: This is a fee paid for the management of the funds. For Oppenheimer, it is 0.63% for all classes of shares. Other Expenses: The other operating expenses were 0.13% for Class A shares and 0.15% for B and C shares. Thus, the total operating expenses for this fund is 1.01% for its Class A shares, and 1.78% for B and C shares. Oppenheimer converts Class A shares to Class B shares after 6 years, so expenses for B shares are 1.01% after the 6th year. 12 Some Additional Notes on Calculation of Expenses and Loads Back-End Load/Contingent Deferred Sales Load: (1) It is calculated as the lesser of the amount that represents a specified percentage of NAV at the time of purchase, or at the time of redemption. (2) It is not applied on shares purchased through reinvestment of dividends or capital gains distributions. (3) It is calculated as if shares that are not subject to a load are redeemed first. (4) Shares are redeemed in the order purchased, unless some other order can result in a lower redemption fee. Operating Expenses: It is applied daily as fraction of NAV. 13 Impact of Costs on Investment Performance (1/5) Let us calculate the impact of the fees on the investor's return. We will use the Oppenheimer growth fund as an example. Consider an investor who starts with $10,000, and can choose between investing in either A, B or C class of shares. Suppose the investor expects that the fund will earn an average of 15% return every year, before expenses. Which class of shares should he invest in? Let us calculate the net return to the investor after costs for different investment horizons. 14 Impact of Costs on Investment Performance (2/5) Class A : 1-Year Horizon Front End Load of 5.75%, total operating expenses 1.01% (12b-1 fee of 0.25%, management fee of 0.63%, other operating expenses of 0.13%). Original investment = $10,000. Amount invested into fund on 1/1/2000 after front-end load = 10,000(1 - 0.0575)= 9,425. Total return before expenses = 15%. Return after expenses of 1.05% = 15-1.01=13.99%. Value of investment on 12/31/2000 = 9425(1+0.1399)=10,743.56 Net return over 1-year = 7.40%. 15 Impact of Costs on Investment Performance (3/5) Class B : 1-Year Horizon: Back End Load of 5.0%, total operating expenses 1.86%. Original investment = $10,000. Amount invested into fund on 1/1/2000 = $10,000. Total return before expenses = 15%. Return after expenses = 15-1.78=13.22%. Value of investment on 12/31/2000 before back-end load= 10000(1+0.1322)=11,322. If we assume that the backend load is applied to the initial amount of $10,000 Value of investment after back-end load of 5% = 11322 - 0.05x10000 = 10,822 Net return over 1-year = 8.22%. *If we assume that the load applies to the final amount, then the value 16 of the fund will be 11322x(1-0.05)=10756, or you will earn 7.56%. Impact of Costs on Investment Performance (4/5) Class C: 1-Year Horizon: No front end load , total operating expenses 1.78%, backend load of 1% in first year. Original investment = $10,000. Amount invested into fund on 1/1/2000 = $10,000. Total return before expenses = 15%. Return after expenses of 1.05% = 15-1.78=13.22%. Value of investment at year-end before back-end load= 10000(1+0.1322)=11,322. Value of investment after back end load of 1% applied to initial investment* = 11322 - 0.01x10,000 = 11222. Net return over 1-year = 12.22%%. (*If the backend load is applied to ending amount, then the value of the investment is 11322x0.99 = $11,209, so that the net return is 12.09%.). 17 Comparing Performance Across Share Classes Investment Annual ReturnsOperating Expenses Front End Load End Back 100 15.00% Class A Classes B&C 5.75% 5.00% 1.01% 1.78% 4.00% 3.00% 2.00% 1.00% Assumptions: Class B convert to Class A after 6th year. Horizon(Yr) Class A 1 107.435575 3 139.598779 5 181.3907461 6 206.7673115 10 349.0992337 20 1293.053315 Class B Class C 108.22 112.22 142.1340958 145.1340958 185.0440361 186.0440361 210.6390577 210.6390577 355.6361646 346.1238338 1317.265915 1198.017083 18 Yet Another Example Vanguard is a large fund family that is particularly known for its passive funds. However, it also has active funds - see the annual report on Vanguard's large cap growth fund: Vanguard US Growth Fund: Annual Report http://www.vanguard.com/funds/reports/usgrar.pdf Its expenses are lower than average, but so are its returns! Moral: Lower expenses by themselves are not a reason to buy active managed funds. 19 Passive Funds Passive funds have much lower expenses as they are simply trying to replicate an index, and thus do not require costly support staff. Moreover, fund returns relative to the benchmark are very sensitive to expenses, and thus there is additional pressure to keep expenses under control. As an example, let us consider the Vanguard Index Trust 500 Fund (VFINX). Class A Net Assets on 31/1/2002 = $73.2B Management fee = 16 bps (0.16%) Total expenses = 18 bps. Return before taxes: -12.02% (1 yr), 1.06% (3 yrs), 10.66% (5 yrs), 12.84% (10 yrs) The next slide provides a comparison with the S&P 20 500. Yearly Investment Return Vanguard 500 Index Fund Inv Year Ended 2001 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 1990 1989 1988 1987 1986 1985 1984 Capital Income Total Return Return Return -13.11% 1.08% -12.02% -9.95% 0.90% -9.06% 19.70% 1.37% 21.07% 27.00% 1.61% 28.62% 31.11% 2.08% 33.19% 20.53% 2.35% 22.88% 34.35% 3.09% 37.45% -1.51% 2.69% 1.18% 7.06% 2.84% 9.89% 4.45% 2.97% 7.42% 3.94% 26.28% 30.22% -6.84% 3.52% -3.32% 26.67% 4.70% 31.36% 11.55% 4.67% 16.22% 2.27% 2.43% 4.71% 14.04% 4.02% 18.06% 26.09% 5.14% 31.23% 1.54% 4.68% 6.21% S&P 500 Total Return -11.89% -9.10% 21.04% 28.58% 33.36% 22.96% 37.58% 1.32% 10.08% 7.62% 30.47% -3.10% 31.69% 16.61% 5.26% 18.68% 31.75% 6.27% 21 Exchange Traded Funds (ETF) Although passive funds can be bought directly from the fund family, a recent innovation is to list a passive fund as an "Exchange Traded Fund" the fund's shares trade continuously on an exchange. (In principle, ETF can be for both active as well as passive funds.) The fund's price tracks the NAV because the ETF allows for redemptions. Exchange traded funds also have low expenses Barclay's charges about 9 bps (0.09%)! The fund saves on marketing costs, as ETF's are listed on an exchange, and thus can be bought and sold like a regular stock. Mostly traded on the AMEX: http://www.amex.com/indexshares/index_shares_over.stm Examples: Barclay's Ishares, Vanguard's VIPER, SPDRs (S&P's Depository Receipts), WEBS (World Equity Benchmark Shares), QQQ (called "cubes,. track the NASDAQ 100) 22 Vanguard's VIPER VIPER: Vanguard Index Participation Equity Receipts Vanguard Total Stock Market VIPER: Tracks the Wilshire 5000 (Ticker: VTI). Expenses of 15 bps. Although it allows for redemptions at NAV, the price can, at times, differ from the NAV. The next page provides details of how much the price differs from NAV. 23 Premium/Discount Since Inception (through 02/08/2002) Closing Price above or equal to NAV Basis Point Differential* 0 - 24 25 - 49 50 - 74 75 - 100 >100 Total Closing Price below NAV Number of Days 55 9 0 0 1 65 % of Total Days 31% 5% 0% 0% 0% 36% Number of Days 103 7 1 0 0 % of Total Days 58% 3% 0% 0% 0% 111 63% 24 Exercises: Please attempt all numerical exercises from the back of Chapter 4. The SEC has provided a calculator to help investors estimate the total cost over the lifetime of the fund: see http://www.sec.gov/mfcc/getstarted.html 25 Chapter 8 (See Also Chapters 5-7) Basic Portfolio Mathematics 26 Road Map 1. Averaging: Geometric vs Arithmetic. 2. Calculation of Portfolio Returns and Variances. 3. Introduction to Asset Allocation 27 Estimating the Mean Return (1/6) We can estimate the mean return in two ways: Arithmetic Mean and Geometric Mean. Suppose you want to estimate the mean return over the last three years, when the returns were r1, r2,and r3. Arithmetic Average = (r1+r2+r3)/3 Geometric Average = [(1+r1)*(1+r2)*(1+r3)]^(1/3)-1 Note that the above method to calculate the geometric average is better than estimating is as [(r1)(r2)(r3)]^(1/3) 28 Arithmetic Vs Geometric (2/6) Consider the following examples: 1. r1=r2=r3=0.10 Arithmetic average = (0.1+0.1+0.1)/3=0.1 Geometric average = [(1.1)*(1.1)*(1.1)]^(1/3)-1 = 0.1 In this case, when all returns are identical, the arithmetic average is equal to the geometric average. In general, this is not true. 29 Arithmetic Vs Geometric (3/6) 2. r1=0.10, r2=0.15, r3=0.05. Arithmetic average = (0.10+0.15+0.05)/3=0.10. Geometric Average = [(1.10)(1.15)(1.05)]^(1/3)1=0.09924 The arithmetic average is greater than the geometric average. Qt: which average to use? 30 The Difference Between Geometric and Arithmetic Average (4/6) There are two points to note: 1. The arithmetic average return will be always greater than or equal to the geometric average return. 2. The difference between the arithmetic and geometric return will depend on the volatility of the return. The greater the volatility, the greater will be the difference in the return. If the volatility is zero (or the returns in every period are the same) then both averages will be the same.* *Approximately, AA - GA = 0.5 (vol^2) 31 Choice Between Arithmetic and Geometric (5/6) 1. If you are simply trying to predict the next period's return, then the arithmetic average will be, statistically, the better choice. 2. If you are trying to calculate the cumulative return over the past 3-year period, the geometric average is better. For example, the arithmetic average of 0.10 estimates the total 3-year return as (1+0.10)^3-1=33.1%, while the geometric average estimates it as (1+0.09924)^3-1=32.825%. In comparison, the exact three year return is (1.1)(1.15) (1.05)-1=32.825%. Thus, if you know the geometric average, you can recover the cumulative return over the period. However, with the arithmetic average you will over-estimate the cumulative return. 32 Geometric Vs Arithmetic: Past Historical Returns (6/6) The difference between estimates of geometric (GA) and arithmetic average (AA) are quite substantial. Here are some estimates over the period 1926-1996 1. Large Cap:AA=12.5%/yr, GA=10.5%/yr 2. Small Cap: AA=19%/yr, GA=12.6%/yr 33 Volatility and Correlations We have already seen that we can easily estimate the volatility and correlation using Excel functions STDEV and CORREL. The variance is defined as the square of the volatility (or standard deviation). Similar to the case of the returns, it is conventional to express the volatility in an annual basis. Annual Volatility = sqrt(12) [Monthly Volatility]. Annual Volatility = sqrt(260)[Daily Volatility]. For example, a daily volatility of 1% implies an annual volatility of about 16%. Recently, we have been observing daily fluctuations of about 1.5% - what does that imply about the annual volatility? 34 Portfolio Return and Variance Suppose we have two assets, with weights w1 and w2, respectively. The weight of w1 is defined as the ratio of the dollar invested in asset 1, divided by the total $ investment. Thus, if you invest $100 in asset 1 and $400 in asset 2, then w1=0.2 and w2=0.8. Portfolio return = w1*r1 + w2*r2 Portfolio variance = (w1*w1)*(var of asset 1) + (w2*w2)*(var of asset2) + 2 (w1)(w2)(correlation)(vol of asset1)(vol of asset 2) To get the portfolio volatility, we take the square root of the portfolio variance. 35 A Digression: On Re-balancing a Portfolio (1/3) We have already observed that we can have different portfolios based on the way we choose the weights - in particular, we saw the equal weighted portfolio and the value weighted portfolio. Qt: How easy is it to maintain such portfolio weights as market prices change? 36 Re-Balancing Cap-Weighted Portfolio (2/3) Company A: #of shares=100, and Price = 20. Company B: # of shares=200, and Price = 40. Therefore: w_a=(100*20)/[100*20+200*40] = 0.20 and w_b=1 - w_a=0.80. Therefore, if you decide to invest a total of $1,000,000 you will buy $200,000 of A ...

Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

UMass (Amherst) - WFCON - 571
Terminology and Fisheries Management Reading: Chapter 1 Review basic definitions Introduce aspects of management issues objectives strategies regulations(Fisheries biology Applied ecology Effects of harvesting on natural communities Empha
UMass (Amherst) - PSYCH - 329000
Child PsychologyPsychology 350Self ConceptSelf Concept Conception of self as object-actor in world Self-recognition at 15-18 months About 2 years children recognize thoughts are not the same as actions About 3 years realization that other
UMass (Amherst) - FOMGT - 421
Derivatives and Investment ManagementThe Use of Futures and OptionsHistory of the Chicago Mercantile Exchange The CME started almost a 100 years ago by listing two commodity contracts: on butter and eggs. Consequently, it was known as the Chicago
UMass (Amherst) - PSYCH - 329000
Child PsychologyPsychology 350Peer RelationsInfancyInterest in peers ~ 3mos Approach, touching ~ 6mos Joyous contagion, physical conflict ~ 18mosBeginning of conflict, aggression.Peers and Socialization 3November 28, 2006Preschool
UMass (Amherst) - WFCON - 470
Population A group of organisms of the same species occupying a particular place at a particular time Population boundaries Genetic differences Habitat differences Manage at the population levelPopulation dynamicsEnvironmentProcessesGrowt
UMass (Amherst) - PSYCH - 329000
Chapter ThirteenGender1Sex v. Gender as a LabelSex is a biological term: Refers to the functional differences associated with male and female reproductive potential; also the differences associated with XX vs XY chromosome configuration Mal
UMass (Amherst) - BIOCH - 623
REVIEWSFORMATION AND TRANSFER OF DISULPHIDE BONDS IN LIVING CELLSCarolyn S. Sevier and Chris A. KaiserProtein disulphide bonds are formed in the endoplasmic reticulum of eukaryotic cells and the periplasmic space of prokaryotic cells. The main pa
UMass (Amherst) - BIOCH - 623
Self-dot matrix: same sequenceWindow 15 Threshold 23Window 10 Threshold 23Window 10 Threshold 15Conservation and covariance within pore lining loops1InvariantConservatively substitutedHypervariableBranched amino acids2Estimation
UMass (Amherst) - BIOCH - 623
Kinetics is the study of the rates at which chemical reactions occurPurposegain an understanding of a reaction mechanismThermodynamics vs Kinetics Thermodynamics tell us whether a given process can occur spontaneously A P Keq = [P]/[A] G'0 = -
UMass (Amherst) - BIOCH - 623
Protein-lipid interactions and the lipid raft hypothesisor ?Fivaz & Meyer 2003 Neuron 40, 319-30 Munro 2003 Cell 115, 377-88How to target proteins to membranes?1Covalent lipid modification of proteins Palmitoylation (16 C acyl chain) (Cys)
UMass (Amherst) - BIOCH - 623
Effect of Trp environment on the emission spectra of proteinsIntramolecular interactions and Interaction with other molecules: cooperativity and bindingOverview of Fluorescence Spectroscopy. FRETPrinciples of Fluorescence Spectroscopy 3rd. Ed. L
UMass (Amherst) - BIOCH - 623
Spectral Methods of Characterizing Protein C f P t i Conformation and ti d Conformational Changes Absorption SpectroscopyObservation/Determination of Structure Chemical techniques: change in the chemical structure Making and breaking structure. of
UMass (Amherst) - BIOCH - 623
Kinetics is the study of the rates at which chemical reactions occurEnergetics and Dynamics of Binding Protein-ligand interaction: affinity Whether or not any particular affinity of a protein for a ligand is significant depends on the concentrat
UMass (Amherst) - BIOCH - 623
Proteins Domains, Motifs, and Quaternary StructurePetsko & Ringe 1-5 to 1-18 Voet-Voet Chapter 8-1 to 8-3 2004 New Science Press Ltd new-science-press.com 2004 New Science Press Ltd new-science-press.com 2004 New Science Press Ltd new-science-
UMass (Amherst) - BIOCH - 623
Enzyme Catalysis Catalysis and l d control of protein function To understand catalysis we must first appreciate the important distinction i t th i t t di ti ti between reaction equilibria and reaction rates The function of a catalyst is to INCREASE
UMass (Amherst) - BIOCH - 623
Spectral Methods of Characterizing Protein C f P t i Conformation and ti d Conformational Changes CD and Fluorescence SpectroscopyCircular Dichroism One only gets a CD signal when two conditions are both fulfilled: you are looking at an asymmetri
UMass (Amherst) - BIOCH - 623
X-ray crystal structures of ColicinsModel of colicin membrane binding and pore formationClosed stateOpen state (little structural information available)FRET analysis between Trp and Cys-labeled DansylModel from FRET data1How to reduced
UMass (Amherst) - BIOCH - 623
2/7/2008Polymeric organization of proteins, nucleic acids, and polysaccarides.ProteinsNucleic Acids (DNA)Glycogen, StarchBranched Structure of GlycogenThe reaction mechanism of glycogen phosphorylase.12/7/2008Reaction catalyzed by U
UMass (Amherst) - BIOCH - 623
Biochemistry 623 ADVANCED GENERAL BIOCHEMISTRY Spring 2008Course Instructors: Dan Hebert, Biochem. & Mol. Biol. LGRT 914, 545-3122, dhebert@biochem.umass.edu Alejandro P. Heuck , Biochem. & Mol. Biol., LGRT 816, apheuck@biochem.umass.edu Classes: in
UMass (Amherst) - BIOCH - 623
TIBS 25 DECEMBER 2000REVIEWS11 12 filament length in striated muscle: rulers and caps or dynamic instability? Annu. Rev. Cell Biol. 14, 487525 Trinick, J. and Tskhovrebova, L. (1999) Titin: a molecular control freak. Trends Cell Biol. 9, 377380 P
UMass (Amherst) - BIOEP - 540
Summarizing Data, Part 1: Summarizing Data with GraphicsWhy summarizing data? It is difficult to learn anything from raw data The part of Statistics that deals with summarizing data is growing more and more important, given the large amount of inf
UMass (Amherst) - BIOEP - 540
Biostatistics serves two purposes (among others): To use information from a sample of data to 1. Describe our best guess of the characteristics of the population. estimation Best guess 2. Gauge the plausibility of alternative explanations for what i
UMass (Amherst) - BIOEP - 540
So far Estimation More Applications defined appropriate confidence interval estimates for a single population mean, . Confidence interval estimators are valuable because they provide: Indicate the width of the central (1-alpha)% of the samplin
UMass (Amherst) - BIOEP - 540
University of Massachusetts: School of Public Health and Health Sciences BioEpi 540-Fall 2004 Take Home Portion of Final ExamAppendix to Solution- 12/6/2004 12:10:07 PM -Welcome to Minitab, press F1 for help.Results for: MINIDAT[1].MTW Histogr
UMass (Amherst) - BIOEP - 540
options pagesize=55 linesize=72 nodate nonumber nocenter nofmterr;*;* Program Documentation Information ; %LET prg=ini01p01.sas; * Name of SAS Program ; TITLE1 "SOURCE: &prg on 09/24/01 by Initials
UMass (Amherst) - SOM - 640
Section 8.4 Reading Bond ListingsProblem. Given bond prices and yields as published by the financial press or other information sources, obtain the U.S. Treasury Yield Curve. Solution Strategy. Collect information about Treasury Bills and Treasury S
UMass (Amherst) - AY - 0203
2002-3 FSUC&ECC Meeting times and places (scheduled) Time PlaceTh Sept 19, 13:30-15:30 Rm 911-13 Campus CenterTh Oct 31, 12:00-13:30 Rm 917 Campus CenterTh Nov 21, 12:00-14:00 Rm 905-09 Campus Cent
UMass (Amherst) - AY - 9900
FSUC&ECC Meeting times and places (scheduled) Time PlaceTu Dec 5, 12:30-13:30 Rm 174-176 Campus CenterW Feb 7, 12:30-13:30 Rm 174-176 Campus CenterF Mar 9, 12:30-13:30 Rm 174-176 Campus CenterTu Apr 10, 1
UMass (Amherst) - BIOEP - 740
Source:bem01p07.sas BE740 4/26/01 EJSTable 1. Example of Data Generated by Simulation with &nubj Subjects each measured at 4 Times where the overall mean is 46 subjects are randomly selected, subject effects are normal with var=292 th
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=90 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem03p20.SAS BE740 4/10/
UMass (Amherst) - BIOEP - 740
Source:bem01p02.SAS BE740 4/12/2001 EJSThe CONTENTS ProcedureData Set Name: WORK.BASE Observations: 291Member Type: DATA Variables: 13Engine: V8
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=90 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem03p16.SAS BE740 3/27/
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=90 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem03p11.SAS BE740 3/11/
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=90 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem03p25.SAS BE740 4/22/
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=110 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem04p08.SAS BE740 5/11
UMass (Amherst) - BIOEP - 740
Source:bem01p17.sas BE740 5/14/01 EJSTable 1. Example with Longitudinal Slope=.5 cross-sectional Slope=-.5 Ave Age Age Ob
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=90 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem03p22.SAS BE740 4/10/
UMass (Amherst) - BIOEP - 740
Source:bem01p18.SAS BE740 5/15/2001 EJSTable 1. Contents of weat3 Data setThe CONTENTS ProcedureData Set Name: WORK.WEAT3 Observations: 1218Member Type: DATA Variables:
UMass (Amherst) - BIOEP - 740
Source:bem01p09.sas BE740 4/26/01 EJSTable 1. Example of Data Generated by Simulation with 52 Subjects each measured at 4 Times where the overall mean is 46 subjects are randomly selected, subject effects are normal with var=292 the r
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=110 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem03p30.SAS BE740 5/06
UMass (Amherst) - BIOEP - 740
OPTIONS LINESIZE=90 PAGESIZE=55 NOCENTER NODATE NONUMBER NOFMTERR;*;* BioEpi 740 Longitudinal Data Analysis ;* PROGRAM NAME LOCATION DATE PROGRAMMER ;TITLE1 "Source:bem03p18.SAS BE740 3/27/
UMass (Amherst) - BIOEP - 740
Source:bem01p03.SAS BE740 4/19/2001 EJSTable 1. Contents of Weather Data set: First four days for each of 40 cThe CONTENTS ProcedureData Set Name: WORK.WEAT1 Observations: 160Member Type: DATA
UMass (Amherst) - BIOEP - 740
Source:bem01p13.sas BE740 5/03/01 EJSExample of created times Obs ID CA2 ageg QUARTER t2 BDATE TC 1 1 Male 35-49 1 1.00000 04/27/96 184 2 1
UMass (Amherst) - BIOEP - 740
Source:bem01p04.SAS BE740 4/24/2001 EJSTable 1. Contents of weat2 Data set for 1996The CONTENTS ProcedureData Set Name: WORK.WEAT2 Observations: 366Member Type: DATA V
UMass (Amherst) - BE - 640
Highlights 14 BE640 Non-Parametric Tests in SAS General Points Aimed at Testing hypotheses- not estimation Simple ideas Require few(er) assumptions. May be nearly as good as parametric tests. (sometimes better) Non-Para metrics Using SAS Binomial Tes
UMass (Amherst) - BE - 640
Final Exam BioEpi 640 Spring 2004 Intermediate BiostatisticsName: _Please answer all questions on these papers. Output for answering the questions is given at the end of the exam. Data for this exam is taken from a portion of the subjects For all
UMass (Amherst) - BE - 640
Final Exam Grade Dist: BioEpi 640 Freq: Spring 2003 Freq: Intermediate Biostatistics90s xxx80s 70s 60s 50s xxxxx xxxxx xxxxx xxx xx xx x Name: _SOLUTION_Please answer all questions on these papers. Output for answering the questions is given at
UMass (Amherst) - BE - 640
BioEpi 640 Intermediate Biostatistics Exam 2- SOLUTION April 19, 2005 Please answer all questions, showing all work on the paper. Good Luck. 1. Two factors are under study for a medication to lower diastolic blood pressure (DBP): method of drug deliv
UMass (Amherst) - BE - 640
Results for Exam on Anova/Reg from Seasons Study Introduction We have abstracted data from the Seasons study. The data set (ex204.sas7bdat) and program (sea04p05.sas) are contained on the Resources-Data page of the course WEB site. You may view each
UMass (Amherst) - BE - 640
Exam 1 BioEpi 640 Intermediate Biostatistics Name: _SOLUTION_ Please answer all questions on these papers (each part of each question is worth 5 pts). 1. Body mass index is computed by taking a persons weight (in Kg) and dividing by the square of the
UMass (Amherst) - BE - 640
Assignment 9 Reading Chapter 22 (pp639-646) on Maximum Likelihood in Kleinbaum, Kupper, et al. Chapter 23 (pp656-664) Logistic Regression Analysis. in Kleinbaum, Kupper, et al. Review Programs in esb05p27.sas; esb05p28.sasProblems (Note: Any SAS out
UMass (Amherst) - BE - 640
Assignment 8. Mixed Models Solutions Reading: Reading: (46 pages) Text: Text: Kleinbaum, Kupper, Muller, and Nizam (1998) 3rd Edition, Applied regression Analysis and Other Multivariable Methods. Repeated measures/Mixed models Chapter 18. (Randomized
UMass (Amherst) - BE - 640
Highlights 12 BE640 Logistic RegressionA. Method of Maximum Likelihood (Chapter 22. Kleinbaum, Kupper, Muller and Nizam) (See also: esb05p27.sas for example of plots of Likelihood function) For Binomial L = P (Y | n, ) n ( n Y ) = Y (1 ) Y =
UMass (Amherst) - BE - 640
Assignment 5. Multiple Linear Regression Applications Solutions Reading: (43 pages) Text: Text: Kleinbaum, Kupper, Muller, and Nizam (1998) 3rd Edition, Applied regression Analysis and Other Multivariable Methods. Chapter 10. (Correlation) pp160-165.
UMass (Amherst) - BE - 640
Highlights 1 Terminology Population Parameter Parameterizations Representation of a deterministic model Random Variable Distribution Statistic Representation of a Stochastic model Dependent variable (response variable) Independent variable (predictor
UMass (Amherst) - BE - 640
Example of How to Compute a Binomial Probability Using MinitabExample (from Daniels, 6th ed., page 90) Suppose that it is known that in a certain population 10% of the population is colorblind. If a random sample of 25 people is drawn from this popu
UMass (Amherst) - BE - 640
Assignment 8 Reading Chapter 14, Klinebaum et al. (pp317-332) (Dummy variables in Regression) Computing Problems1.Run the analyses using the program from the WEB esb640p06.sas and the data set CF2.sas7bdat. Write up a 2 page report that summarizes th
UMass (Amherst) - BE - 640
Assignment 2 Reading Kleinbaum, Kupper, Muller, and Nizam (1998). Applied regression analysis and multivariable methods, Duxbury Press. Chapter 4+5, pp 34-60 Group Computing Problem (write up 1-2 pages and a computer program that can be used to selec
UMass (Amherst) - BE - 640
Final Exam BioEpi 640 Spring 2004 Computer Output PROGRAM:esb64p04p17.sas on May 15, 2004 Table 1. Contents of Data set to describe persons Who Change their diet The CONTENTS Procedure Data Set Name: WORK.D1 Observations: 443 Member Type: DATA Variab
UMass (Amherst) - BE - 640
LOGISTIC REGRESSION ANALYSIS GOAL: To find the best fitting, simplest, model possible describing the relationship between an outcome (dependent or response) variable and a set of independent (predictor or explanatory) variables.or "covariates".Wha
UMass (Amherst) - BE - 2002
115 99 92181 83 75181 92 74128 64 74107 130 9484 66 7976 89 89118 54 7396 80 61110 76 62110 84 60 62 67 67