Options appear in the order they were printed on the original paper, not by their position on the page. 99 of the 100 printed questions are here; questions 14 (figure-based, without a verified key, or held back in review) are left out.
Diagram in which area of the rectangle is proportional to the frequency of the class intervals with unequal length
Histogram
Algebraic sum of the deviations of a set of values from their arithmetic mean is :
Zero
If the distribution is moderately assymetrical then 3(mean - median) is equal to
mean - mode
While comparing two groups it is observed that coefficient of variation of one group is 2 along with mean as 4, while in other group coefficient of variation is 3 and mean is 8 . The ratio of two standard deviations is :
1/3
A given data has mean = 6.5, median = 6.3 and mode = 5.4 . It represents :
positively skewed distribution
Complete idea of a distribution is formed with :
All of these
Let correlation coefficient between X and Y is positive then the sign of both the regression coefficient is :
positive
Let X; (i =1,2,3) are uncorrelated variables each having the same standard deviation, then correlation coefficient between (X₁ + X₂) and (X2 + X3) is:
1/2
For a discrete distribution for obtaining rank correlation coefficient, the sum of all the differences of ranks that is, Σ di i=1 is:
Zero
If the regression equation is a good fit then the multiple correlation coefficient R is -
near to 1
Correlation coefficient between variables X₁ and X2, X₁ and X3, and X2 and X3 is r, then partial correlation coefficient between X₁ and X2 after eliminating the effect of X3 is :
r r+1
Minimizing the sum of squares of the deviation of the actual value of Y from their estimated value as given by the line of best fit in a linear model is :
Principle of Least squares
For any three events A, B and C P(A U BIC), is equal to
P(AC) + P(B|C) - P(A ∩ B|C)
Let A and B be two events such that P(A) = 1/3, P(B)= 1/5 and P (AUB) = 11/30 then P(B|A) is :
1/2
Let two unbiased coins are tossed. Expected value of the number of heads is :
1
It is observed that P(Ej) = 1/3 for i = 1 and 2, and P(B|E₁) = 4/5, Baye's probability P(E2|B) is -
2/3
Let X be a random variable with p.d.f. as given below x: 0 1 2 3 p(x): 1/3 1/2 1/24 1/8 The expected value of Y = (X-1)² is
7/8
The moment generating function of Binomial distribution is :
(q + pet)n
Let f(x;r,p) denote probability distribution function of negative Binomial distribution. If r = 1 then this distribution becomes
Geometric
In normal probability distribution all central moments of odd order are
zero
Mean and variance of exponential distribution are respectively
1/λ and 1/λ²
In which distribution probability density function at all the points through out the range of X is same
Rectangular
Let X and Y be independent Gamma variates with parameters µ and y, then distribution of Z = X/(X+Y) is :-
Beta distribution of type-I
Let X ~ ẞ2 (m, n), then y = 1/(1+X) follows
Beta distribution of type-I
Let X ~ F (n,m) then distribution of 1/X follows :-
F (m,n) distribution
Let X ~ χ²(n) and Y ~ χ²(m) are two independent variates, then the ratio of these two chi squares distributions divided by their respective degrees of freedom follows :-
F-distribution
Which one is not true for the assumption of t-test?
population variance is known
In an experiment it is observed that d = 4,n = 9 and s² = 36, then testing the significance of difference between two means, paired t-test comes out as
2
Characteristic function of χ²- distribution with n degrees of freedom is
(1 - 2it)⁻ⁿ/²
Let statistic t follows student t-distribution with n degrees of freedom. If n=1 then this distribution reduced to
Standard cauchy distribution
Mode of F - distribution is always
less than 1
Let X follows p-variate normal distribution with mean vector μ and variance covariance matrix Σ, that is, X~Nₚ(μ, Σ), then Y = cX follows
Nₚ(cμ, cΣc')
Wishart distribution is a generalization of
χ²- distribution
If Aᵢ (i = 1,2,....,p) are independently distributed as Wₚ(Σ, ηᵢ) respectively, then A = ∑ᵢ<0xE2><0x82><0x91>₁ᵖ Aᵢ is distributed as
Wₚ(Σ, Σᵢ<0xE2><0x82><0x91>₁ᵖ ηᵢ)
The test statistics t<0xE2><0x82><0x99>₋<0xE2><0x82><0x96>₋₁ for testing the multivariate regression parameters βⱼ under null hypothesis H₀: βⱼ = 0 is
bj / Sbj
For perfect multicollinearity, which one is not true
|X'X| is not equal to zero
Which one of the following statement in not correct :
The generalized variance of the vector of principal component is not the generalized variance of the original vector.
For small samples from N(θ,σ²); σ² unknown a 100(1 - α) percent confidence interval for θ is given by
θ̂ ± t<0xE2><0x82><0x90>₋₁ SE (θ̂)
For a sample of size n, the statistic s² = 1/n Σ_{i=1}^{n} (xᵢ − x̄)² is
biased and consistent estimator
An M.V.U. estimator is unique in the sense that if T₁ and T₂ are M.V.U. estimators for γ (θ) then
T₁ = T₂ almost surely
In an experiment it is observed that E(X) = θ and I(θ) = n/θ², then Cramer-Rao lower bound for the variance of an unbiased estimator of θ is
θ²/n
Let the likelihood function of a density function is given by L(a) = f(x, a) = (a - x); 0 < x < a then maximum likelihood estimate of parameter a is
2x
If t<0xE2><0x82><0x99> is a sufficient estimator for θ, then ∂/∂θ log L is a function of only:
t<0xE2><0x82><0x99> and θ
Which one is not true
We may reject Ho when it is false, is type-I error
Which one is not true:
α is the probability of committing type-ll error
For testing H₀: θ = θ₀ against H₁: θ = θ₁, the critical region W and consequently, the test of size α based on it said to be unbiased if the power of the test is
≥ α
If the direction of the difference is not known, the critical region will be in both ends of the sampling distribution, the test is known as
two tailed test
If f(x, θ) = 1/θ; 0 ≤ x ≤ θ 0 otherwise For testing the null hypothesis H₀: θ = 1 against H₁: θ = 2 base on a single observation x the size of type-l error if we choose critical region as 0.5 ≤ x is
0.50
If T = t(x) is sufficient statistics for θ then using N-P lemma, the most powerful critical region for the test may be defined in terms of
Marginal distribution of T = t(x)
In a sign test r is the number of positive sign, then the distribution of r is
Binomial
If the sample size in Wald Wolfovitz runs test are n₁=25 and n₂=30 then the variate R (total number of runs) has mean
311/11
In a nonparametric median test the test criteria follows the distribution
Hypergoemetric
For Kolmogorov-Smirnov one sample test, which one is not true
D<0xE2><0x82><0x99> = |S<0xE2><0x82><0x99>(x) - F(x)|
For Kolmogorov-Smirnov two sample test, which one is not true
Empirical distribution is compared with population cumulative distribution function
In case of sign test for paired samples, the hypothesis is to test the
difference of median
The error mainly arising at the stage of ascertainment and processing of data is
non- sampling error
A sample of size 2 from a population of size 5 is drwan using simple random sampling, the probability of selecting one sample is
1/10
When the procedure of selecting units in the stratra is specified then which of the following statement is NOT correct for the procedure of stratification?
none of these
If 'k' is sampling interval then in systematic sampling, sample mean ỹ is unbiased estimator for population mean Ỹ, if
C) N = nk
Which sampling method is a particular case of PPS sampling when P₁ = 1/N, i = 1,...., N
C) Simple random sampling
The approximate expression for the bias in the ratio estimator becomes zero if R is equal to -
A) cov(X,Y) / V(X)
Ratio estimator YR of population mean is more efficient than regression estimator Y₁ of population mean when :
A) p ≥ Rx Sy / Sx
A 24 factorial experiment is confounded into a block of size 4. Total number of generalized interaction(s) is / are
A) 1
The number of replication 'r' is determined with usual notations from the following:
A) r = 2t² s²/d²
Experimental error is due to:
A) extraneous factors
A 23 factorial experiment is conducted in 3 replications. Degrees of freedom for error sum of squares is :
D) 14
For a balanced incomplete block design, the parametric relations are (1) λ(V – 1) = r(k - 1) (2) b ≥ V (3) r = k (4) Vr = bk Which one is true for the existence of BIBD ?
D) 1,2,4
A partially balanced incomplete block design satisfy the following inequalities (1) r - λ₁ < 0 (2) r - λ₁ > 0 (3) rk – υλ₂ < 0 (4) rk – υλ₂ > 0 Which one is true for regular group divisible PBIBD ?
D) 2,4
For which design two way local control holds true
C) LSD
The price of certain commodities for four years is computed. It is observed that quartely average of first quarter is 360 and over all average of four quarters is 320. Seasonal index of first quarter is -
A) 112.5
In cyclic variation, the objective is to
C) eliminate trend and eliminate seasonal effect from time series data
In a time series, Harmonic analysis method is based on the function yt expressed in the form of :
C) Fourier series
Under usual assumption in time series data y₁ = Bo + B1 Yt-1 + Et is a type of
B) Auto regressive model
The price of two commodities A and B in the year 2015 and 2016 are shown in Table. The average price relative of both commodities is assuming 2015 as base year
A) 1.45
Price Index, for a year using simple aggregate method, Pon, for various commodities is
B) ΣPn / ΣPo * 100
The main advantages of Laspeyre's index number is that the weights (qo) for the base year
remains the same throughout
For a time series data index number obtained from chain base method is expressed by: (where j = 1,2,....,k ; t = 1,2,....,n)
Σpjt q(j-1)t / Σρ(j-1)t q(j-1)t
Paasche's index number possess :
downward bias
Using base shifting method, the new base index number is obtained from
old index number of current year / index number of new base year * 100
Estimation of population, for a particular year using Analytical method, is obtained from
Po + n/N (P1 - Po)
Fertility rates mainly depend upon :
female population in the child bearing age
Number of children born alive per 1000 women of child bearing age in a given year and region is
General fertility rate
Which of the following are the assumptions of Gross Reproduction Rate (GRR) are (1) There is no mortality of newly born female children till they attain the highest reproductive age. (2) There are no gains or loses due to migration. (3) The current fertility rate is maintained till their highest child bearing age. Which one is most appropriate for GRR ?
(1), (2) and (3)
If total fertility rate is 1050 and the ratio of female birth to male is 100 : 110 then gross reproductive rate is :
500
With usual notation which of the below given relation is not correct :
Lx = x + 1/2 x
In which year planning cell was set up in CSO ?
1955
NSSO is an autonomous body since the year:
1971
In which year first census of free India was conducted ?
1951
Which one of the following method is not true for the calculation of National income ?
Survey Method
Functioning of CSO is: (A) to co-ordinate the activities of statistical officer at the centre and state. (B) to estimate the yearly income of the country.
both (A) and (B)
Agricultural Statistics is mostly collected by : (A) Department of Economics and Statistics at central level. (B) Department of Economics and Statistics at state level.
both (A) and (B)
Which statement is not correct:
there is only one possible alleles at each locus.
The data refers to the index numbers of Agricultural production with base year as the triennium end given details regarding
food grains and non food grains
Probability of including a specified unit in a sample of size 4 out of 20 units is :
1/5
The graph between the period (µ) of cycle and intensity (Su) is known as:
periodogram
If r12 = 0.6, r13 = 0.5 and r23 = 0.8 then the value of r12.3 is:
0.38
If the probability of occurence of two independent events A and B respectively are 0.3 and 0.4, then which of the following is true :
P (A and B ) = 0.12
If X ~ N (μ, Σ) then with usual notations T² is defined as:
T² = (x – μο)'S⁻¹ (x – μο) / 1/N
The unbiased estimator of population variance σ² is : Where s² = 1/n Σ(xi - x)²
n/(n-1) * s²
Where you lost marks — practise 10 questions on each:
| Subject | Questions |
|---|---|
| Probability & Distributions | 24 |
| Applied Statistics | 23 |
| Statistical Inference | 22 |
| Sampling & Design of Experiments | 13 |
| Correlation & Regression | 8 |
| Descriptive Statistics | 7 |
| Economics | 1 |
Yes. It is the paper from the 2015 sitting conducted by RPSC — 99 questions (of the 100 printed) in the order they were printed, with the answer key matched to the board's final key.
The official answer key is on this page — pick an option and the right answer shows, or use "Show the full answer key" for all of them. The solution to each question shows after a free Google sign-in.
Yes, the whole paper is free. All 99 questions can be read without paying anything.
The printed paper had 100 questions; 99 are here. Questions 14 are left out — ones that depend on a figure, have no verified answer key, or were held back in review. Printing a question without its figure or a sure answer would be wrong.
Probability & Distributions — 24 questions. The full count per subject is in the table above.