Scribd Logo
Upload

Welcome to Scribd!

  • 408normal PDF

    Uploaded by

    OMSURYACHANDRAN
    7 views2 pages

    Original Title

    408normal.pdf

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Report this Document

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    7 views2 pages

    408normal PDF

    Uploaded by

    OMSURYACHANDRAN

    Copyright:

    © All Rights Reserved
    Download as PDF or read online from Scribd
    Flag for inappropriate content
    of 2
    Math 408, Actuarial Statistics I A.J. Hildebrand The Normal Distribution —x/% on ob, fi) e mit! ‘The #-funetion a Te I o 2 lt u * Definition: ©(z) is the area under the “Bell curve function” (1//2)e~*"/? between 00 to z, ie, ®(2) = Ze [2,,e-* Pat (see picture. * Probabilistic significance: (2) is the cumulative distribution function (o.4.£.) of the standard normal distribution (see below). « Properties: () ®(-00) = 0, (00) = 1, ®(2) is increasing; (i) &(-2) = 1- &(2), ®@) =0. Properties (i) are just general properties of any ¢.d.f,; properties (ii) express the symme- try of ® with respect to the y-axis. ‘* Normal table: Since there exists no “explicit” formula for ®(z) (the integral repre- senting (zr) cannot be evaluated in terms of elementary functions), for computations involving © one has to resort to tabulated values of ®(c). Such a “normal table” can be found in the back of Hogg/Tanis, and will be provided in actuarial exams. Using calculators with built-in ® function, or with integrating capabilities, is not allowed in actuarial exams. ‘+ Upper percentiles: Upper percentiles (or “upper percent points”) are defined just like ordinary percentiles, but with respect to the complementary probability. For example, the 5th upper percentile is the z-value for which 1 — (2) equals 0.05; it is denoted by 205 and approximately equal to 1.64. The 5th upper percentile is the same as the 95th ordinary percentile defined as the z-value at which ®(z) = 0.95. Math 408, Actuarial Statistics I A.J. Hildebrand ‘The standard normal distribution N(0,1) + Cumulative distribution function (cd.f.): F(2) defined above. (x), where & is the function Density (p.d.f.): f(x) = Jgzexp(-}32*) (-00 <2 < 00). Expectation and variance: E(X) =0, Var(X) =1 Moment-generating function: M(t) = exp (2) Computation of probabilities: If X is standard normal N(0,1), then Pla< X a) =1-H(a). The general normal distribution N(, 07) * Cumulative distribution function (c.d-f. defined above. F( 4% (54), where ® is the function 1 # Density (p.d.f.): f(c) = =Jpz exp (- HY) (-00< <0) © Expectation and variance: E(X) = 1, Var(X) = * Moment-generating function: M(t) = exp (jut + 407¢?). (Note that the factor 0? here is in the numerator, not in the denominator, as in the formula for the pdt, f(z).) « Standardizing a normal random variable: Rescaling a normally distributed rv. (with general values ofp: and ) by subtracting its mean and dividing by its standard deviation (ie., 0, not 0) gives one that has standard normal distribution. In other words, if X is N(u,02), then Z = (X ~ p)/o is N(0,1). ‘* Computation of probabilities: To compute probabilities involving a rv. X’ with normal distribution N (1,02), first convert the probabilities into probabilities involving the standardized version of X, Z = (X — p)/o, then express the latter probabilities via the ® function as above. For example, pa

    You might also like