ST903: Statistical Methods Assignment 1
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
ST903: Statistical Methods
Assignment 1
1: An company produces wooden planks for decorative panelling. It is established by their quality
control process that the mean length is equal to the design value of 3 m and the standard deviation is 3 cm.
(i) Without making any assumptions about the distribution, give an upper bound on the probability that a randomly selected plank has a length above 315 cm? [2]
(ii) How large a standard deviation is acceptable if the probability of observing a plank above
330 cm may not be larger than 0.0001? [2]
(iii) Suppose the distribution of lengths can be assumed to be is Gaussian. How large a
standard deviation is acceptable if the probability of observing a plank above 330 cm may not be larger than 0.0001? [2]
(iv) Would either of the previous two answers be changed if a plank below 270 cm also needed
to be avoided? [2]
(v) What do the previous answers tell us about the usefulness of distributional assumptions in statistics? [2]
[TOTAL: 10]
2: (i) Let X and Y be random variables with finite means. Show that
minE[(Y − g(X))2] = E[(Y − E[Y |X])2]
g(x)
where g(x) ranges over all functions. E[Y |X] is sometimes called the regression of Y on
X, the “best” predictor for Y conditional on X . [3]
(ii) Let X1 and X2 be independent standard Normal random variables, i.e. Xi ∼ N(0, 1),i =
1, 2. Find the distribution of the random variable . [3]
[TOTAL: 6]
3: Let X and Y be independent Poisson distributed random variables with parameters θ and λ ,
respectively i.e. X ∼ Poi(θ) and Y ∼ Poi(λ).
(i) Write down the joint density function of (X,Y), indicating the possible values X and Y can take. [1]
(U,V), indicating clearly which values U and V can take. [4]
(iii) Derive the marginal density function of U . [4]
(iv) Interpret the result found in (iii) and state your finding as a theorem. [2]
(v) Derive the conditional density function of V |U and state which known distribution it
corresponds to. [3]
[TOTAL: 14]
2022-11-04