REDEEMER’S UNIVERSITY, EDE TEXTS: FUNDAMENTALS OF STATISTICS BY S.C GUPTA 7 TH EDITION STATISTICS AND ECONOMETRICS BY DOMINICK SALVATORE SCHAUM’S OUTLINE SERIES IN ECONOMICS. COURSE DESCRIPTION Probability Theory Random variables and their Distributions Special Probability Models ( Discrete and Continuous) Sampling and Sampling Distribution Estimation Hypothesis Testing Special Test Statistics and Procedure Introduction to Statistical Research in Economics CONTINUED… o Research Design and Surveys in Economics o Collection and Analysis of Secondary Data o Designing Questionnaire o Analysis of Qualitative and Quantitative Data o Non Parametric Statistical Data. PROBABILITY THEORY Introduction: - Definition of Terms - Notation - Sampling Approaches to Probability - Classical approach - Empirical approach Certain an Impossible events - Rules of probability Theory Introduction Definition of Terms: - Outcome: In any trial, each separate possible result is called an outcome - Event: The particular occurrence looked being looked for in a trial is the event. - Success: when the event occurs, we have a success. - Failure: When it does not occur, we have a failure. Continued… In general terms, probability is a measure of the likelihood that a particular event will occur in any one trial, or experiment, carried out in prescribed conditions. The ability to predict likely occurrences has obvious applications e.g. in insurance matters and in industrial quality control and the efficient use of resources. Notation… The probability that a certain event A will occur is denoted by P(A). For example, if A represents the event that a component, picked at random stock, is faulty, written A={faulty component}, then P(A) denotes the probability of picking a faulty component. Sampling In a manufacturing run, it would be both time-consuming and uneconomical to subject every single component produced to full inspection. It is usual, therefore, to examine a sample batch of components, taken at random, as being representative of the whole output. The larger the random sample, the more nearly representative of the whole population (total output of components) is likely to be. Types of Probability The determination of probability may be taken from two approaches: (a) empirical ( or experimental) probability (b) classical ( or theoretical) probability Let us look at each of these in turn. Empirical Probability Empirical probability is based on previous known results. The relative frequency of the number of times the event has previously occurred is taken as an indication of likely occurrences in the future. Example, a random batch of 240 components is subjected to strict inspection and 20 items are found to be defective. Therefore, if we pick any one component at random from this sample, the chance of its being faulty P ( A) nA ‘’20 in 240; i.e. ‘1 in 12’ . This is given by N Continued…. Where P( A) = probability that event A will occur n A = no. of ways that event A can occur N = total no. of equally possible outcome P A 0 If event A is an impossible or null event or probability P A 1 of failure. If , event A will occur with certainty. Continued… So, if A ={faulty components} Then P(A)=20/240 or 1 in 12 =1/12= 0.0833=8.33% The most usual forms are P(A)=1/12 or P(A) = 0.0833 Exercise: Assume a run of 600 components from the same machine. What would be the likely number of defectives? Ans. 50 Expectation The result does not assert that there will be exactly 50 defectives in any run of 600 components. But, having found the probability of the event occurring in any one trial (1/12), we can use it to predict the likely number of times the event will occur in N similar trails. This expectation E is defined by the product of the number of trials and the probability P(A) that the event A will occur in any one trial, i.e E N P( A) Success or failure Throughout, we are concerned with the probability of the occurrence of a particular event. When it does occur in any one trial, we record a success . When it fails to occur, we record a failure – whatever, the defined event may be. If, in N trials, there are x successes, there will also be (N- x) failures, so that x+(N-x)=N Therefore, x/N+N-x/N=1 Continued…. But, x/N=P(success)=P(A) and N-x/N=P(failure)=P(not A) Therefore P(A)+P(not A) = 1 The event (not A) is called the complement of event A and is often written as A, i.e. P( A) P( A ) 1 Therefore, If the probability of picking a defective in any one trial is 1/5, the probability of not picking a defective would be what? Continued… Example: Fifteen percent of castings are found to be outside prescribed tolerances. Determine the number of acceptable items likely to be present in a batch of 120 such castings. Given that 15 % are rejects, Hence, 85 % acceptable. Therefore, if event A ={acceptable casting}, P(A) =85/100 =17/20 N = 120 and E = N x P(A) = 120 x 17/20 = 102 i.e. expected number acceptable. Practice Question It is known that from past records that 8% of moulded plastic items are defective. Determine (a) the probability that any one item is (i) defective (ii) acceptable (b) the number of acceptable items likely to be found in a sample batch of 4500 Classical Probability The classical approach to probability is based on a consideration of the theoretical number of ways in which it is possible for an event A to occur. The classical probability P of an event A occuring is defined by: P(A)= no. of ways in which event A can occur/ total no. of all possible outcomes Example: If we consider the chance result of rolling a normal unbiased die, the total no. of possible outcomes is six i.e. 1, 2, 3, 4, 5, 6. Continued…. If the event we are considering is the throwing of a 6, this can occur in only one way out of the six possible outcomes. Therefore, P(Six)=1/6; p(not six)=5/6 If out of n possible outcomes of a trial, it is possible for an event A to occur in x ways and for an event A not to occur in y ways, then n= x+y and P(A)=x/n = x/x+y Certain and Impossible Events (a) If an event is certain to occur every time, then x=n, y = 0 . Therefore, P(A) =n/n=1 (b) If an event A cannot possibly occur at any time, then x=0, y=n Therefore; P(A)=0/n = 0 i.e. P(certainty)=1; P(impossibility)=0 In most cases, probability values lie between the extreme values. So- if one card is drawn at random from a full pack of playing cards, the probability of obtaining (a) a heart=13/52=1/4 (b) a king 4/52=1/13 (c) a card other than a king =48/52=12/13 (d) a black card? In a deck where half of the cards are red and half are black, the probability of drawing a green card is zero or an impossible event. Probability Rules In the earlier section, probability was computed as a ratio or a weight assigned to the occurrence of an event. However, quite often we are interested in the probability of the occurrence of more complex events. For example, we may want to find the probability that a king or a club will occur in a draw from a deck of 52 cards. Similarly, on examining couples with two children, if one child is known to be a boy, we may be interested in the prob. of the event of both the children being boys. To compute the probabilities of these kind, probability rules are required. Mutually exclusive and Mutually non- exclusive events Mutually exclusive events are events which cannot occur together. For example, in rolling a die, the event of throwing a six and that of throwing a five cannot occur at the same time. Similarly, when drawing a card from a pack, the event of drawing an ace and also that of drawing a king cannot occur in the same single trial Mutually non-exclusive events are events that can occur simultaneously. For example in rolling a die, the event of obtaining a multiple of 3 and the event of obtaining a multiple of 2 can occur together if a six is thrown. Similarly, with cards, the event queen of clubs; queen of spades can occur together. Addition Rule of Probability Given that events A and B are mutually exclusive, the probability P of either event A or event B occurring- but clearly not both - is P (A or B) = P(A ) +P(B) If the events A and B are non-exclusive, so that events A and B can occur together, then the probability of A or B occurring is given by P (A or B)= P(A) +P(B)-P(AnB) Example: Mutually Exclusive Events A single card is drawn from a pack of 52 playing cards. What is the probability of drawing either an ace or a seven? Let event A={drawing an ace} and event B={drawing a seven} The probability o drawing either an ace or a seven, i.e. P(A or B)= P(A)+P(B) 4 aces in the pack, Therefore, P(A) =4/52=1/13 4 sevens in the pack, therefore P(B)=4/52=1/13 Hence, P(A or B)= P (ace or seven)=1/13+1/13=2/13 Ans. Example: Non exclusive events In rolling a die, what is the probability of scoring a multiple of 3 or a multiple of 2? Multiple of 3 (i.e 3 0r 6) and multiple of 2 (2, 4, 6,) = P (A) +P(B) P(A)=1/6+1/6=2/6 P(B)= 1/6+1/6+1/6=3/6 Therefore, the probability of scoring a multiple of 3 or a multiple of 2 = P(A)+P(B)= 2/6+3/6 = 5/6 Note: One outcome (6) appears in both sets of outcomes, ( i.e. not mutually exclusive), this outcome has been counted twice. The prob. of scoring one of the sixes must therefore be removed. P(six)=1/6=P(AnB), hence, P(A or B)=2/3 Ans. Practice Questions A card is drawn from a well-shuffled pack of playing cards. Find the probability that the card drawn is either a club or a king. Suppose your chance of being offered a certain job is 0.45, your probability of getting another job is 0.55, and your probability of being offered both jobs is 0.30. What is the probability that you will be offered at least one of the two jobs? Independent and dependent events Independent events: Events are independent when the occurrence of one event does not affect the probability of the occurrence of the second event. Dependent Events: Events are dependent when one event does affect the probability of the occurrence of the second.