Document Sample

Stats 845 Applied Statistics This Course will cover: 1. Regression – Non Linear Regression – Multiple Regression 2. Analysis of Variance and Experimental Design The Emphasis will be on: 1. Learning Techniques through example: 2. Use of common statistical packages. • SPSS • Minitab • SAS • SPlus What is Statistics? It is the major mathematical tool of scientific inference - the art of drawing conclusion from data. Data that is to some extent corrupted by some component of random variation (random noise) An analogy can be drawn to data that is affected by random components of variation to signals that are corrupted by noise. Quite often sounds that are heard or received by some radio receiver can be thought of as signals with superimposed noise. The objective in signal theory is to extract the signal from the received sound (i.e. remove the noise to the greatest extent possible). The same is true in data analysis. Example A: Suppose we are comparing the effect of three different diets on weight loss. An observation on weight loss can be thought of as being made up of two components: 1. A component due to the effect of the diet being applied to the subject (the signal) 2. A random component due to other factors affecting weight loss not considered (initial weight of the subject, sex of the subject, metabolic makeup of the subject.) random noise. Note: that random assignment of subjects to diets will ensure that this component will be a random effect. Example B In this example we again are comparing the effect of three diets on weight gain. Subjects are randomly divided into three groups. Diets are randomly distributed amongst the groups. Measurements on weight gain are taken at the following times - - one month - two months - 6 months and - 1 year after commencement of the diet. In addition to both the factors Time and Diet effecting weight gain there are two random sources of variation (noise) - between subject variation and - within subject variation This can be illustrated in a schematic fashion as follows: Deterministic factors Diet Time Response weight gain Random Noise within subject between subject The circle of Research Questions arise about a phenomenon Conclusion are drawn A decision is made to from the analysis collect data Statistics Statistics A decision is made as The data is how to collect the summarized and data analyzed The data is collected Notice the two points on the circle where statistics plays an important role: 1.The analysis of the collected data. 2.The design of a data collection procedure The analysis of the collected data. • This of course is the traditional use of statistics. • Note that if the data collection procedure is well thought out and well designed, the analysis step of the research project will be straightforward. • Usually experimental designs are chosen with the statistical analysis already in mind. • Thus the strategy for the analysis is usually decided upon when any study is designed. • It is a dangerous practice to select the form of analysis after the data has been collected ( the choice may to favour certain pre- determined conclusions and therefore in a considerable loss in objectivity ) • Sometimes however a decision to use a specific type of analysis has to be made after the data has been collected (It was overlooked at the design stage) The design of a data collection procedure • the importance of statistics is quite often ignored at this stage. • It is important that the data collection procedure will eventually result in answers to the research questions. • And will result in the most accurate answers for the resources available to research team. • Note the success of a research project should not depend on the answers that it comes up with but the accuracy of the answers. • This fact is usually an indicator of a valuable research project.. Some definitions important to Statistics A population: this is the complete collection of subjects (objects) that are of interest in the study. There may be (and frequently are) more than one in which case a major objective is that of comparison. A case (elementary sampling unit): This is an individual unit (subject) of the population. A variable: a measurement or type of measurement that is made on each individual case in the population. Types of variables Some variables may be measured on a numerical scale while others are measured on a categorical scale. The nature of the variables has a great influence on which analysis will be used. . For Variables measured on a numerical scale the measurements will be numbers. Ex: Age, Weight, Systolic Blood Pressure For Variables measured on a categorical scale the measurements will be categories. Ex: Sex, Religion, Heart Disease Types of variables In addition some variables are labeled as dependent variables and some variables are labeled as independent variables. This usually depends on the objectives of the analysis. Dependent variables are output or response variables while the independent variables are the input variables or factors. Usually one is interested in determining equations that describe how the dependent variables are affected by the independent variables A sample: Is a subset of the population Types of Samples different types of samples are determined by how the sample is selected. Convenience Samples In a convenience sample the subjects that are most convenient to the researcher are selected as objects in the sample. This is not a very good procedure for inferential Statistical Analysis but is useful for exploratory preliminary work. Quota samples In quota samples subjects are chosen conveniently until quotas are met for different subgroups of the population. This also is useful for exploratory preliminary work. Random Samples Random samples of a given size are selected in such that all possible samples of that size have the same probability of being selected. Convenience Samples and Quota samples are useful for preliminary studies. It is however difficult to assess the accuracy of estimates based on this type of sampling scheme. Sometimes however one has to be satisfied with a convenience sample and assume that it is equivalent to a random sampling procedure A population statistic (parameter): Any quantity computed from the values of variables for the entire population. A sample statistic: Any quantity computed from the values of variables for the cases in the sample. Statistical Decision Making • Almost all problems in statistics can be formulated as a problem of making a decision . • That is given some data observed from some phenomena, a decision will have to be made about the phenomena Decisions are generally broken into two types: • Estimation decisions and • Hypothesis Testing decisions. Probability Theory plays a very important role in these decisions and the assessment of error made by these decisions Definition: A random variable X is a numerical quantity that is determined by the outcome of a random experiment Example : An individual is selected at random from a population and X = the weight of the individual The probability distribution of a random variable (continuous) is describe by: its probability density curve f(x). i.e. a curve which has the following properties : • 1. f(x) is always positive. • 2. The total are under the curve f(x) is one. • 3. The area under the curve f(x) between a and b is the probability that X lies between the two values. 0.025 0.02 0.015 f(x) 0.01 0.005 0 0 20 40 60 80 100 120 Examples of some important Univariate distributions 1.The Normal distribution A common probability density curve is the “Normal” density curve - symmetric and bell shaped Comment: If m = 0 and s = 1 the distribution is called the standard normal distribution 0.03 Normal distribution 0.025 with m = 50 and s =15 0.02 Normal distribution with m = 70 and s =20 0.015 0.01 0.005 0 0 20 40 60 80 100 120 xm 2 1 2 f(x) e 2s 2s 2.The Chi-squared distribution with n degrees of freedom 1 (n 2 ) / 2 x / 2 f ( x) n n / 2 x e if x 0 2 2 0.5 0.4 0.3 0.2 0.1 2 4 6 8 10 12 14 Comment: If z1, z2, ..., zn are independent random variables each having a standard normal distribution then U = z1 z2 zn 2 2 2 has a chi-squared distribution with n degrees of freedom. 3. The F distribution with n1 degrees of freedom in the numerator and n2 degrees of freedom in the denominator n 1 n 2 / 2 n1 f(x) K x x (n1 2)2 1 if x 0 n 2 n1 / 2 n1 n 2 n1 2 n2 where K = n1 n2 2 2 0.8 0.7 0.6 0.5 F dist 0.4 0.3 0.2 0.1 0 0 1 2 3 4 5 6 Comment: If U1 and U2 are independent random variables each having Chi-squared distribution with n1 and n2 degrees of freedom respectively then U 1 n1 F= U 2 n2 has a F distribution with n1 degrees of freedom in the numerator and n2 degrees of freedom in the denominator 4.The t distribution with n degrees of freedom n1 / 2 x 2 f(x) K 1 n n 1 2 where K = n n 2 0.4 0.3 0.2 0.1 -4 -2 2 4 Comment: If z and U are independent random variables, and z has a standard Normal distribution while U has a Chi- squared distribution with n degrees of freedom then z t= U n has a t distribution with n degrees of freedom. • An Applet showing critical values and tail probabilities for various distributions 1. Standard Normal 2. T distribution 3. Chi-square distribution 4. Gamma distribution 5. F distribution The Sampling distribution of a statistic A random sample from a probability distribution, with density function f(x) is a collection of n independent random variables, x1, x2, ...,xn with a probability distribution described by f(x). If for example we collect a random sample of individuals from a population and – measure some variable X for each of those individuals, – the n measurements x1, x2, ...,xn will form a set of n independent random variables with a probability distribution equivalent to the distribution of X across the population. A statistic T is any quantity computed from the random observations x1, x2, ...,xn. • Any statistic will necessarily be also a random variable and therefore will have a probability distribution described by some probability density function fT(t). • This distribution is called the sampling distribution of the statistic T. • This distribution is very important if one is using this statistic in a statistical analysis. • It is used to assess the accuracy of a statistic if it is used as an estimator. • It is used to determine thresholds for acceptance and rejection if it is used for Hypothesis testing. Some examples of Sampling distributions of statistics Distribution of the sample mean for a sample from a Normal popululation Let x1, x2, ...,xn is a sample from a normal population with mean m and standard deviation s Let x i x i n Than x i x i n has a normal sampling distribution with mean mx m and standard deviation sx s n 0 20 40 60 80 100 Distribution of the z statistic Let x1, x2, ...,xn is a sample from a normal population with mean m and standard deviation s Let xm z s n Then z has a standard normal distibution Comment: Many statistics T have a normal distribution with mean mT and standard deviation sT. Then T mT z sT will have a standard normal distribution. Distribution of the c2 statistic for sample variance Let x1, x2, ...,xn is a sample from a normal population with mean m and standard deviation s Let xi x 2 s2 i = sample variance n 1 and xi x 2 s i = sample standard deviation n 1 Let x x 2 i (n 1)s 2 c 2 i s 2 s 2 Then c2 has chi-squared distribution with n = n-1 degrees of freedom. 0 .5 The chi-squared distribution 0 0 4 8 12 16 20 24 Distribution of the t statistic Let x1, x2, ...,xn is a sample from a normal population with mean m and standard deviation s Let xm t s n then t has student’s t distribution with n = n-1 degrees of freedom Comment: If an estimator T has a normal distribution with mean mT and standard deviation sT. If sT is an estimatior of sT based on n degrees of freedom Then TmT t sT will have student’s t distribution with n degrees of freedom. . t distribution standard normal distribution Point estimation • A statistic T is called an estimator of the parameter q if its value is used as an estimate of the parameter q. • The performance of an estimator T will be determined by how “close” the sampling distribution of T is to the parameter, q, being estimated. • An estimator T is called an unbiased estimator of q if mT, the mean of the sampling distribution of T satisfies mT = q. • This implies that in the long run the average value of T is q. • An estimator T is called the Minimum Variance Unbiased estimator of q if T is an unbiased estimator and it has the smallest standard error sT amongst all unbiased estimators of q. • If the sampling distribution of T is normal, the standard error of T is extremely important. It completely describes the variability of the estimator T. Interval Estimation (confidence intervals) • Point estimators give only single values as an estimate. There is no indication of the accuracy of the estimate. • The accuracy can sometimes be measured and shown by displaying the standard error of the estimate. • There is however a better way. • Using the idea of confidence interval estimates • The unknown parameter is estimated with a range of values that have a given probability of capturing the parameter being estimated. Confidence Intervals • The interval TL to TU is called a (1 - a) 100 % confidence interval for the parameter q, if the probability that q lies in the range TL to TU is equal to 1 - a. • Here , TL to TU , are – statistics – random numerical quantities calculated from the data. Examples Confidence interval for the mean of a Normal population (based on the z statistic). s s TL x z a / 2 to TU x z a / 2 n n is a (1 - a) 100 % confidence interval for m, the mean of a normal population. Here za/2 is the upper a/2 100 % percentage point of the standard normal distribution. More generally if T is an unbiased estimator of the parameter q and has a normal sampling distribution with known standard error sT then TL T z a / 2 sT to TU T za / 2s T is a (1 - a) 100 % confidence interval for q. Confidence interval for the mean of a Normal population (based on the t statistic). s s TL x t a / 2 to TU x t a / 2 n n is a (1 - a) 100 % confidence interval for m, the mean of a normal population. Here ta/2 is the upper a/2 100 % percentage point of the Student’s t distribution with n = n-1 degrees of freedom. More generally if T is an unbiased estimator of the parameter q and has a normal sampling distribution with estmated standard error sT, based on n degrees of freedom, then TL T t a / 2s T to TU T t a / 2s T is a (1 - a) 100 % confidence interval for q. Common Confidence intervals Situation Confidence interval Sample form the Normal distribution with unknown s0 mean and known variance x za / 2 (Estimating m) (n large) n Sample form the Normal distribution with unknown s mean and unknown variance (Estimating m)(n small) x ta / 2 n Estimation of a binomial probability p p (1 p ) ˆ ˆ p za / 2 ˆ n Two independent samples from the Normal 2 2 distribution with unknown means and known sx s y variances x y za / 2 (Estimating m1 - m2) (n,m large) n m Two independent samples from the Normal 1 1 distribution with unknown means and unknown but x y ta / 2 s Pooled equal variances. (Estimating m1 - m2) ) (n,m small) n m Estimation of a the difference between two binomial p1 (1 p1 ) p2 (1 p2 ) ˆ ˆ ˆ ˆ probabilities, p1-p2 p1 p2 za / 2 ˆ ˆ n1 n2 Multiple Confidence intervals In many situations one is interested in estimating not only a single parameter, q, but a collection of parameters, q1, q2, q3, ... . A collection of intervals, TL1 to TU1, TL2 to TU2, TL3 to TU3, ... are called a set of (1 - a) 100 % multiple confidence intervals if the probability that all the intervals capture their respective parameters is 1 - a Hypothesis Testing • Another important area of statistical inference is that of Hypothesis Testing. • In this situation one has a statement (Hypothesis) about the parameter(s) of the distributions being sampled and one is interested in deciding whether the statement is true or false. • In fact there are two hypotheses – The Null Hypothesis (H0) and – the Alternative Hypothesis (HA). • A decision will be made either to – Accept H0 (Reject HA) or to – Reject H0 (Accept HA). The following table gives the different possibilities for the decision and the different possibilities for the correctness of the decision • The following table gives the different possibilities for the decision and the different possibilities for the correctness of the decision Accept H0 Reject H0 H0 Correct Type I is true Decision error H0 Type II Correct is false error Decision • Type I error - The Null Hypothesis H0 is rejected when it is true. • The probability that a decision procedure makes a type I error is denoted by a, and is sometimes called the significance level of the test. • Common significance levels that are used are a = .05 and a = .01 • Type II error - The Null Hypothesis H0 is accepted when it is false. • The probability that a decision procedure makes a type II error is denoted by b. • The probability 1 - b is called the Power of the test and is the probability that the decision procedure correctly rejects a false Null Hypothesis. A statistical test is defined by • 1. Choosing a statistic for making the decision to Accept or Reject H0. This statisitic is called the test statistic. • 2. Dividing the set of possible values of the test statistic into two regions - an Acceptance and Critical Region. • If upon collection of the data and evaluation of the test statistic, its value lies in the Acceptance Region, a decision is made to accept the Null Hypothesis H0. • If upon collection of the data and evaluation of the test statistic, its value lies in the Critical Region, a decision is made to reject the Null Hypothesis H0. • The probability of a type I error, a, is usually set at a predefined level by choosing the critical thresholds (boundaries between the Acceptance and Critical Regions) appropriately. • The probability of a type II error, b, is decreased (and the power of the test, 1 - b, is increased) by 1. Choosing the “best” test statistic. 2. Selecting the most efficient experimental design. 3. Increasing the amount of information (usually by increasing the sample sizes involved) that the decision is based. Some common Tests Situation Test Statistic H0 HA Critical Region n x m0 m m m m Sample form the Normal distribution with unknown z < -za/2 or z > za/2 z mean and known variance s m m z > za (Testing m) (n large) m m z <-za n x m 0 m m m m Sample form the Normal distribution with unknown t < -ta/2 or t > ta/2 t mean and unknown variance (Testing m) (n small) s m m t > ta m m t < -ta Testing of a binomial p p0 ˆ p p p p z < -za/2 or z > za/2 probability p z p0 (1 p0 ) p p z > za n p p z < -za Two independent samples from the Normal distribution z x y m1 m 2 m1 m 2 z < -za/2 or z > za/2 with unknown means and 2 2 known variances sx s y (Testing m1 - m2) m 1 m 2 z > za (n, m largel) n m m1 m 2 z < -za Two independent samples from the Normal distribution t x y m1 m 2 m1 m 2 t < -ta/2 or t > ta/2 with unknown means and 1 1 unknown but equal s Pooled m 1 m 2 t > ta variances. (Testing m1 - m2) n m m1 m 2 t < -ta Estimation of a the p1 p 2 ˆ ˆ p1 p 2 p1 p2 z < -za/2 or z > za/2 difference between two z binomial probabilities, p1-p2 1 1 p (1 p) ˆ ˆ n1 n2 p1 p 2 z > za p1 p2 z < -za The p-value approach to Hypothesis Testing In hypothesis testing we need 1. A test statistic 2. A Critical and Acceptance region for the test statistic The Critical Region is set up under the sampling distribution of the test statistic. Area = a (0.05 or 0.01) above the critical region. The critical region may be one tailed or two tailed The Critical region: a/2 a/2 za / 2 0 za / 2 z Reject H0 Reject H0 Accept H0 PAccept H 0 when true P za / 2 z za / 2 1 a PReject H 0 when true Pz za / 2 or z za / 2 a In test is carried out by 1. Computing the value of the test statistic 2. Making the decision a. Reject if the value is in the Critical region and b. Accept if the value is in the Acceptance region. The value of the test statistic may be in the Acceptance region but close to being in the Critical region, or The it may be in the Critical region but close to being in the Acceptance region. To measure this we compute the p-value. Definition – Once the test statistic has been computed form the data the p-value is defined to be: p-value = P[the test statistic is as or more extreme than the observed value of the test statistic] more extreme means giving stronger evidence to rejecting H0 Example – Suppose we are using the z –test for the mean m of a normal population and a = 0.05. Z0.025 = 1.960 Thus the critical region is to reject H0 if Z < -1.960 or Z > 1.960 . Suppose the z = 2.3, then we reject H0 p-value = P[the test statistic is as or more extreme than the observed value of the test statistic] = P [ z > 2.3] + P[z < -2.3] = 0.0107 + 0.0107 = 0.0214 Graph p - value -2.3 2.3 If the value of z = 1.2, then we accept H0 p-value = P[the test statistic is as or more extreme than the observed value of the test statistic] = P [ z > 1.2] + P[z < -1.2] = 0.1151 + 0.1151 = 0.2302 23.02% chance that the test statistic is as or more extreme than 1.2. Fairly high, hence 1.2 is not very extreme Graph p - value -1.2 1.2 Properties of the p -value 1. If the p-value is small (<0.05 or 0.01) H0 should be rejected. 2. The p-value measures the plausibility of H0. 3. If the test is two tailed the p-value should be two tailed. 4. If the test is one tailed the p-value should be one tailed. 5. It is customary to report p-values when reporting the results. This gives the reader some idea of the strength of the evidence for rejecting H0 Multiple testing Quite often one is interested in performing collection (family) of tests of hypotheses. 1. H0,1 versus HA,1. 2. H0,2 versus HA,2. 3. H0,3 versus HA,3. etc. • Let a* denote the probability that at least one type I error is made in the collection of tests that are performed. • The value of a*, the family type I error rate, can be considerably larger than a, the type I error rate of each individual test. • The value of the family error rate, a*, can be controlled by altering the thresholds of each individual test appropriately. • A testing procedure of this nature is called a Multiple testing procedure. A chart illustrating Statistical Procedures Independent variables Dependent Categorical Continuous Continuous & Categorical Variables Multiway frequency Analysis Discriminant Analysis Discriminant Analysis Categorical (Log Linear Model) ANOVA (single dep var) MULTIPLE ANACOVA Continuous MANOVA (Mult dep var) REGRESSION (single dep variable) (single dep var) MULTIVARIATE MANACOVA MULTIPLE (Mult dep var) REGRESSION (multiple dependent variable) Continuous & ?? ?? ?? Categorical Next topic: Fitting equations to data Link

DOCUMENT INFO

Shared By:

Categories:

Tags:
Page Views, Art Icons, Maximum Size, desktop icons, PNG icons, windows icons, mac icons, Real Estate, the Society, Last Week

Stats:

views: | 5 |

posted: | 4/14/2011 |

language: | English |

pages: | 114 |

OTHER DOCS BY gjjur4356

How are you planning on using Docstoc?
BUSINESS
PERSONAL

By registering with docstoc.com you agree to our
privacy policy and
terms of service, and to receive content and offer notifications.

Docstoc is the premier online destination to start and grow small businesses. It hosts the best quality and widest selection of professional documents (over 20 million) and resources including expert videos, articles and productivity tools to make every small business better.

Search or Browse for any specific document or resource you need for your business. Or explore our curated resources for Starting a Business, Growing a Business or for Professional Development.

Feel free to Contact Us with any questions you might have.