oe vs | INTRODUCTION Te sry STATISTICAL THEORY Part 1 (A text book for Degree and Post-Graduate Students) By Prof. Sher Muhammad Chaudhry B.Sc. (Hons.), M.A. (Gold Medalist) F.8.S. (London) Ex-Head, Department of Statistics Government College (now GC University), Lahore Dr. Shahid Kamal . M.Sc. (Gold Medalist), Ph.D. (U.K.) Principal and Professor College of Statistical & Actuarial Sciences University of the Punjab, Lahore ILMI KITAB KHANA® Kabir Street, Urdu Bazar, Lahore 54000. ~~“ | \ (Pakistan) ge an agar Scanned with CamScannerv Preface CONTENTS ‘ TRODUCTION wy 21 2.2 23 we 25 26 27 Meaning of Stati LL” Use of al Information 1.1.2 Characteristics of Statistics 1.13 Descriptive and Inferential Statistics 1.1.4 Populations and Samples 1.1.5 Importance of Statistics Observations and Variables Variables. Discrete and Continuous Variables Measurement Scales Errors of Measurement Significant Digits Rounding off a Number Collection of Data 13.1 Collection of Primary Data 1.3.2 Coliection of Secondary Data 1.3.3. Editing of Data 13.4 Uses and Misuses of Statistics Exercises PRESENTATION OF DATA Introduction Classification 2.2.1 Aims of Classification 2.2.2 Basic Principles of Classification Tabulation 23.1 Types of Tables 23.2. Main Parts of a Table and its Construction Frequency Distribution 2.4.1 Class-limits 2.4.2 Class-boundaries 2.4.3. Class-Mark 2.4.4 Class Width or Interval 2.4.5 Constructing a Grouped Frequency Distribution 2.4.6 Cumulative Frequency Distribution Stem-and-Leaf Display Graphical Representation Diagrams 2.7.1 Simple Bar Chart 2.7.2 Multiple Bar Chart 2.7.3 Component Bar Chart Page COMOIAHVUHa RYN 10 ll i oT 15 1S 15 1S 1S 16 16 20 20 20 20 20 21 26 27 28 29 30 30 31 Scanned with CamScannerXK we Rectangles and Sub-divided Rectangles Pictograms Pie Diagrams Profit and Loss Chart we a Graph of Time Series ~ Historigram Histogram Frequency Polygon Frequency Curve Cumulative Frequency Polygon or Ogive Ogive for a Discrete Variable ‘Types of Frequency Curves Ratio Charts or Semi-logarithmic Graphs Exercises MEASURES OF CENTRAL TENDENCY OR AVERAGES 3.1 Introduction Criteria of a ‘Types of Averag The Arithmetic Mean 3.4.1 The Weighted Arithmetic Mean 3.4.2 Properties of the Arithmetic Mean 3.4.3 Mean from Grouped Data 3.4.4 Change of Origin and Scale 3.5 The Geometric Mean 3.6 The Harmonic Mean 3.7 The Median 3.7.1 Quantiles 3.8 The Mode 3.9 Empirical Relation between Mean, Median and Mode 3.10 The Box Plots 3.11 Relative Merits and Demerits of Various Averages 3.11.1 The Arithmetic Mean 3.11.2 The Geometric Mean 3.11.3 The Harmonic Mean 3.11.4 The Median 3.11.5 The Mode Exercises factory Average MEASURES OF DISPERSION, MOMENTS AND SKEWNESS Al Introduction 4.2 Range 43 The Semi-Interquartile Range or the Quartile Deviation 4.4 The Mean (or Average) Deviation Scanned with CamScanner 32 33 44 35 35 36 37 39 40 40 41 41 43 44 5s 55 By 56 57 37 59 61 162 64 o7 68 2 4 14 15 15 1 16 1 1% 1 87 87 88 . 945 ene Standard Deviation oe of Origin and Scale . interpretation of the S - 453 Co-efficient anda Deviation 4.5.4 Properties of Varian , 45.5 Standardized ai Standard Deviation 46 Trimmed and Winsorized Measures 4.7 Moments 471 Moments about the Mean in terms of Moments about an arbitrary origin, say a, and conversely 4.7.2 Sheppard’s Corrections 4.7.3. Moment-Ratios 4.7.4 Change of Origin and Scale 4.7.5 Charlier Check 48 — Skewness 49 Kurtosis 4.10 Describing a Frequency Distribution Exercises ae) INDEX NUMBERS 5.1 Introduction 5.1.1 Simple and Composite Index Numbers 5.1.2 Problems Involved in Index Number Construction 5.2 Main Steps in the Construction of Index Numbers of Wholesale Prices 5.2.1 Selection of Commodities for Inclusion 5.2.2 _ Selection of the Base Period 5.2.3 Selection of Average 5.2.4 Selection of Appropriate Weights 5.3. Unweighted Index Numbers 5.3.1 Simple Ageregative Index 5.3.2 Simple Average of Relatives 5.4. Weighted Index Numbers 5.4.1. Weighted Aggregative Price Index Numbers 542 Weighted Average of Relatives Price Index Number 5.5 Quantity Index Numbers 5.6 ‘Tests for Index Number Formulae 5.6.1 Time Reversal Test 5.6.2 Factor Reversal Test 5.6.3 Circular Test 5.7 Consumer Price Index Number 5.7.1 Meaning 5.7.2 Construction of Consumer Price Index Numbers ol 96 97 98 100 103 104 105 106 108 108 110 Ml 114 115 116 116 131 131 131 132 132 132 134 135 135 135 136 140 140 145 147 151 151 153 155 156 156 156 Scanned with CamScannery 5.8 59 5.7.3. Shortcomings or Drawbacks of Consumer Prig Numbers rice Index Uses of Index Numbers. imitations of Index Numbers. ‘xerci 6. PROBABILITY 6.1 AD ‘ Introduction An Aside — Sets Subsets. Venn Diagram Operations on Sets The Algebra of Sets Partition of Sets. Class of Sets Cartesian Product Sets 6.2.8 Relation and Function Random Experiment 6.3.1 Sample Space 6.3.2 Events 6.3.3 Events and Symbolic Representations 63.4 Counting Sample Points Definitions of Probability 6.4.1 Subjective or Personalistic Probability Laws of Probability Conditional Probability Independent and Dependent Events Exercises 7. RANDOM VARIABLES L OP vAD 4@ 75 7.6 Introduction Distribution Function Discrete Random Variables and its Probability Distribution Continuous Random Variable and its Probab lity Density Function Joint Distributions 75.1 Bivariate Distribution Function Bivariate Probability Functions Marginal Probability Functions Conditional Probability Functions Independence 7.5.6 Continuous Bivariate Distributions Mathematical Expectation of a Random Variable 7.6.1 Expectation of a Function of a Random Variable Scanned with CamScanner 159 160 160 161 173 21 27 228 233 231 237 238 238 2B 239 243 248 250INTRODUCTION PATISTICS People view Statistics in many different ways. Generally it is considered to be a subject that deals with percentages, charts, graphs, averages and tables. Some people think that Statistics is a subject consisting of rules, methods and techniques of collecting and presenting large amount of numerical information, while other people think that it is a subject of making inferences about the population on the basis of sample information, The word “Statistics” ‘h comes from the Latin word status, meaning a political state, originally meant information useful to the state, for example, information about the sizes of populations and armed forces. But this word has now acquired different meanings. 4 In the first place, the word statistics refers to “numerical facts systematically arranged”. In this sense, the word statistics is always used in the plural. We have, for instonce, statistics of prices, statistics ofroad accidents, statistics of crimes, statistics of births, statistics of educational institutions, etc. In all these examples, the word statistics denotes a set of numerical data in the respective fields. This is the meaning the man in the street gives to the word Statistics and most people usually use the word dara instead fely Example 1.1 In the following examples, the facts and figures usually called Statistics presented in the media almost every day are given: % i) Children who brush their teeth with brand XYZ toothpaste have 60% fewer countries, ii) The Bureau of census projects the population of Pakistan to be 170.1 million in the year 2010. iii) Eight out of ten Pakistanis do not have skills. iv) The prevalence of diabetes is nearly 3 times as high in over weight people 2s compared to normal weight people ¥) In 1980 it was estimated that 0.1% of people had tried any sort of drug; where as in 2008 it was estimated that 10% had done so. ‘In the/second place) the word statistics is defined as a discipline that includes procedures and techniques used to collect, process and analyse numerical data to make inferences and to reach decisions inthe face of uncertainty. It should of course be bome in mind that uncertainty does not imply ignorance but it refers to the incompleteness and the instability of data available, In this sense, the word statistics is wed in the singular. As it embodies more or less all stages of the general process of learning, sometimes called scientific method, statistics is characterized as a science. Thus the word statistics used in the plural. ‘sfers to a set of numerical information and in the singular, denotes the sc of basing de Tumerical data. It should be noted that statistics as a subject is mathematical in character ( diy) the word statistics are numerical quantities calculated from sample observations, a single Auantity that has been so calculated is called a statistic, The mean of a sample for instan * Word statistics is plural when used in this sense. sa statistic 1.1.1 Use of Statistical Information. ‘The statistical information are and can be used for a variety ‘Sons. Some of them are: 4) ~ii) Of rea ‘o inform general public; ‘o explain things that have happened; Scanned with CamScannera i iii) to justify a claim; Ziv) to provide general comparisons; ” vi) toestimate the unknown quantities: sion regarding future outcomes; vil) to establish association / relationship between factors Hence Stausties is a subject which is much more than just numbers. It tells us what is done to g with numbers. The following three examples further & stics may be used xplain how Stat Example 1.2 Suppose we want to determine the best teacher at Govt. College University, Lahore How should we decide this? This could be done by asking Govt. College University stuclents who the bes teacher is. To do so, we collect the-data, analyze the results and make the decision, Now vation questions are: i) should we survey every student? ii) how will the survey be conducted? iii) how will the data be analyzed? iv) how will the best teacher be determined? In order to answer these and other questions, Statistical techniques are used Example 1.3 A TV station claims that an advertisement of @ product on their channel attracts mor cnstomers compared to all other TV channels. Now if this claim is based on data, there it can be used t miaaket the TV channel. Suppose we have some doubts about the claim, In order to remove the doubts, might gather relevant information, analyze the results using appropriate statistical technique and make decision regarding the claim. Example 1.4 Suppose University of the Punjab is planning an expansion program of its physic: facilities. Tu draw up an effective course of action, the University authorities decide that it needs t answer this question, how many college students will we need to accommodate over the next ten yea The question can be further broken dows into many smaller questions, How many college students then be in the Punjab? How many will want to attend the University of the Punjab? ete. Once ag Statistical methods can assist in evaluating and planning of expansion program, 1.1.2 Characteristics of Statistics. The cefinition stated above indicates that statistics is 4 U/® in its own right. It may therefore be desirable to know the characteristic features of statistics in ore" appreciate and understand its genera! nature. Some of its important characteristics are given below i) Statistics deals with the behaviour of aggregates or large groups of data. It has nothing °° with what is happening to a particular individual or object of the aggregate. s ii) Statistics deals with aggregates of ebservations of the same kind rather than isolated fig ob iii) Statistics deals with variability that obscure underlying patterns. No two objects LY universe are exactly alike. If they were, there would have been no statistical problem Scanned with CamScannerPY jpiRODUCTION __ a_— — = _ . in) __ Statsties deals with uncertainties or uncontrolled, involve terms of probability 8 every process of getting observations whether controlled deficiencies or el jeneies or chance variation. ‘That is why we have to talk in 4) Statistics deals with those characteristics. or aspects. of thiny which can be ed numerically either by counts or by measurements, ich can be described ” Satis dea wih hose aggregates which are subject wo » number ofrandom causes, eg the heights of persons are subject to a number of causes such a t i uses such as race, ancestry, age, diet, habits, climate and so forth eee vii) Statistical laws are valid on the average or in the long run, Th no guarantee that a certain law will hold in all cases. Statistical inference is therefore made in the face of uncertainty. vii) Statistical results might be misleading and incorrect if sufficient care in collecting, processing and interpreting the data is not exercised or if the statistical data are handled by a person who is not well versed in the subject matter of statistics. 1.1.3 Descriptive and Inferential Statisties. Statistics as a subject, may be divided into descriptive statistics and inferential statistics. Peseriprive statistics is that branch of statistics which deals with concepts and methods concerned with Summarization and description of the important aspects of numerical data. T! is area of study consists of the condensation of data, their graphical displays and the computation of a few numerical quantities that provide information about the centre of the data and indicate the spread of the observations. sitial statistics) deals with procedures for making inferences about the characteristics that describe the large group of data or the whole, called the population, from the knowledge derived from only apart ofthe data, known as sample. This area includes the estimation of population parameters and testing of statistical hypotheses. This phase of statistics is based on probability theory as the inferences “hich are made on the basis of sample evidence, cannot be absolutely certain. Comparison Descriptive Statistics Inferential Statistics i) Accricket player wants to estimate his chance of scoring based on his current season average. i) A cricket player wants to find his score average for the last 20 games. the ii) Based on the first four test scores, ‘Aamir would like to predict the variation in his final Statistics test scores ii) Aamir wants to describe th variation in his four test scores 1n Statistics. Based on last six months grocery bills, Mrs. Rashid would like 10 predict the average amount she will spend on groceries for the upcoming year. iii) Mrs. Rashid wants to determine iii) the average weekly amount she spent on groceries in the past 6 months. Scanned with CamScannerINTRODUCTION TO STATISTICAL THEORY ; | 7 1.1.4 Populations and Samples, A\population or a statistical population is a collection or set of all ‘ossible observations whether finite or infinite, relevant to some characteristic of interest. A statistical spulation may be real such as the heights of all college students or hypothetical such as all the possible outcomes from the toss of a coin, The number of observations in a finite population is called the size of the population and is denoted by the letter N. Numerical quantities describing a population are called (parameters) customarily represented by Greek letters. It is important to note that ih statistics the word ‘population is a technical term not necessarily referring to all the people in a specified area, rather denoting the aggregate of measurements or counts of some characteristic for the entire group of objects or individuals. 4 7 A fample is a part or a subset of a population. Generally it consists of some of the observations but in certain situations, it may include the whole of the population. The number of observations included in a sample is called the size of the sample and is denoted by the letter 1. A numerical quantity computed from a sample, is called alstatistic, which is usually represented by ordinary Latin letter. The information derived from sample data is used to draw conclusions about the population. Example 1.5 State whether each of the following is a population or a sample, i) Total number of absentees by all students in a college during the last month, ii) Number of colour TV sets owned by all families in Lahore. iii) Monthly salaries of all employees of a company. iv) Wheat yield per acre for 5 pieces of land. v) Number of computers sold during the last month at all the computer stores in Lahore. Solution i) Population ii) Population iii) Population iv) Sample v) Population 1.1.5 Importance of Statistics. Statistics is perhaps a subject that is used by everybody. The following functions and uses of statistics in most diverse fields serve to indicate its importance, i) Statistics assists in summarizing the larger sets of data in a form that is easily understandable. ii) Statistics assists in the efficient design of laboratory and field experiments as well as surveys. iii) Statistics assists in a sound and effective planning in any field of inquiry iv) Statistics assists in drawing general conclusions and in making predictions of how much of thing will happen under given conditions. y) Statistical techniques being powerful tools for analysing numerical data, are used in almost every branch of learning. In the biological and physical sciences, Genetics, Agronom ‘Anthropometry, Astronomy, Physics, Geology, etc. are the main areas where statistical techniques have been developed and are increasingly used. Scanned with CamScannerINTRODUCTION ee 5 vi) A businessman, an industrialist work. Banl and a research worker all employ ies and Governments all have thei stical methods in their tistics departments » Insurance compani vii), A modern administrator whether in uli a factual basis for decision Public or private sector, leans on statistical data to provide viii) A politician uses statistics advant, F a ageou c “1 while elucidating the probleme usly to lend support and credence to his arguments handles, A social scier ix) ‘ust Uses statistical methods in various areas of socio-economic life of a nation. It is sometimes said that “a social scientist without an adeg : ‘ juate underst si often like the blind man groping fanding of statistics, is in a dark room for a black cat that is not there”. 12 OBSERVATIONS AND VARIABLES In statistics, an observation often means any sort of numerically recording of information, whether itisa physical measurement such as height or weight; a classification such as heads or tails, or an answer toa question such as yes or no. 1.2.1. Variables. A characteristic that varies with an individual or an object, is called a variable. For example, age is a variable as it varies from person to person. A variable can assume a number of values. The given set of all possible values from which the variable takes on a value, is called its domain. If for a given problem, the domain of a variable contains only one value, then the variable is referred to as constant. Variables may be classified into quantitative and qualitative according to the form of the characteristic of interest. A variable is called a quantitative variable when a characteristic can be expressed numeriéally such as age, weight, income or number of children. On the other hand, if the characteristic is non-numerical such as education, gender, eye-colour, quality, intelligence, poverty, satisfaction, etc. the variable is referred to as a qualitative variable. A. qualitative characteristic is also called an avtribidte. An individual or an object with such a characteristic can be counted or envimersted after having been assigned to one of the several mutually exclusive classes or categories 1.2.2 Discrete and Continuous Variables. A quantitative variable may be classified as cane continuous, Aldiscrete variable is one that can take only a discrete set of integers or Whole numbers, i cI such : ber e a te variable represents cowne data such as the num! is the va by jumps or breaks. A discre' ore re tony the pander of rooms in a house, the number of deaths in an accident, the income sons i , of an individual, etc. variable if it can take on any value--fractional or integer--within.a .rval with all possible values without gaps. A continuous variable he age of a person, the height of a plant, the weight of a [A variable is called ajeontinuous given interval, i.e. its domain is an inte represents\miedsurement_datai such as 1 commodity, the temperature at a place, etc. is generally denoted by some symbol such as oF + countable or measurable, Jaced by & vara aed fr or jtl value of the variable. The subscript For J EP Yand X,or Y, represents that itl number such as 1,2,3,--- When referred to a particular value. Scanned with CamScanner