0% found this document useful (0 votes)
172 views2 pages

Pyq Time Series Analysis

The document is a question paper for a Time Series Analysis course, containing various questions related to autoregressive models, Brown's discounted regression, and forecasting methods. It includes theoretical questions and practical problems requiring calculations and definitions related to time series components. Candidates are instructed to attempt any five questions from the paper.

Uploaded by

panchalansh5400
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
172 views2 pages

Pyq Time Series Analysis

The document is a question paper for a Time Series Analysis course, containing various questions related to autoregressive models, Brown's discounted regression, and forecasting methods. It includes theoretical questions and practical problems requiring calculations and definitions related to time series components. Candidates are instructed to attempt any five questions from the paper.

Uploaded by

panchalansh5400
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

4493 4 [This question paper contains 4 printed pages.

(c) For the autoregressive series yt+2 + clyt+l + pyr = si+2;


Your Roll No..............,

(a2 40, ol. show that t - r, - lSj una Sr. No, of Question Paper: 4493 G
' l+P
hence that I + o. + B is non-negative. (4,5,6) Unique PaperCode : 32377905

Name of the Paper : Time Series Analysis


6 (a) Describe the Brown's discounted regression
Name of the Course : B.Sc. (Hons.) Statistics -
procedure for a steady time series model and show
DSE
that it is equivalent to exponential smoothing.
Semester : V ;2..1.,,,'.
(b) For the model
Duration : 3 Hours Maj.imm vii?Rs : 7s
(1 B)(l - 0.2B)y, : (l - 0.5B)2, t,/
where {2,} is a discrete-time, purely random stru ct io ns for
process such that E(Zl) = 0, Yar(Zr) : o,2 and
In Candidates
t\
I Write your Roll No. on the top iately on re
successive values of Z, are independent, find the
forecasts for one- and two- steps-ahead and of this question paper.
compute the variance of their errors. Hence or 2 Attempt any five questions.
otherwise, show that a recursive expression for
3 All questions carry equal marks.
forecasts three or more steps ahead is given by

iN(h) = 1.2 iN(h - 1) - 0.2 iN(h - 2)

Further, if zN: l, yr.r:4, yN r =3, oj: 2, find (a) Define a time series. Give any two objectives for
analyzing a series.
!'N(2) and the standard error of the corresponding
forecast error. (7,8) (b) Define the decomposition of time series by additive
and multiplicative hypothesis, clearly outlining the
7 Write notes on any two of the following :
assumptions made.
(a) Selection of trend type
(b) Variate difference method
(c) Identify which component of the time series is
mainly responsible for the movement in the
(c) Bayesian forecasting (71/2,7t/z)
following time series :
(1000) P.T.O.
4493 2 4493 3

(i) Decrease in the construction activity during (ii) show that


rainy months.
mtm2...mr [m, ][m,]...[m,]vo = . du{-.,"Lu'*
(ii) A need for increased rice production due [*
to increase in population.
(iii) if m, = 4, mt= 4, and m, = 5, obtain the weights
(iii) A fall in death rate due to scientific of the iterated averages when the above formula
advancement. is approximated by a cubic polynomial. (15)
(iv) Eruption of a volcano. (5,6,4)
4 (a) Given that a time series is composed of the trend,
2. (a) Enumerate three properties of the curve the oscillatory, and the random components.
Discuss the effect of elimination of the trend on
k
ut b<0 other components of the time series.
l + ea+bt '
and describe its different phases with respect to (b) Let a time series X, observed at unit time intervals
time series d ata of annual production of some contain a deterministic sinusoidal component at a
industry. Just ify that the output of an industry known frequency )",
follows the lo gistic trend. X,: Ao + Arcos)"t + A2sin)"t + Zr
where Z, denotes a purely random process.
(b) What do you understand by 'seasonal variations'
Estimate the unknown parameters Ao, A, and
in a time series? Describe the 'Ratio to Trend' A'2. (8'7)
method of computing the indices of seasonal
variations, stating clearly the assumptions made.
5. (a) Define the terms autocorrelation and correlogram
(7Y,,7%)
(b) Given that z, is a purely random process such that
3 If f[ml
m' ' stands for the simple average of 'm' terms E(2,) : 0, V(2,) : o2 and successive values of zl
then are independent, for what values of ),, and ),, is
the second order AR process
(i) prove that
Yr: IrYr r * \zYrzt z,
I [ *'-l -, ]
m d-vo stationary? Obtain the complementary function of
m
Yo
fvo+ , ] yr, if )"f : l/3 and )v,:219.
P.T.O.

You might also like