Random variables and probability distributions
Mathematical methods · 72 questions and templates for CSIR NET Physics, in order from warm-up to top rank. Check each answer, then read the full solution.
New to this topic? Read the lesson first: Random variables and probability distributions.
Loading questions… (they need JavaScript; nothing you type is sent anywhere)
Written questions with solutions
The same questions as a list, so you can print them or read without the practice box. Templates (fresh numbers) appear only in the practice box above.
-
MCQWarm-up
For a Poisson distribution with mean , the variance isAnswer and solution
Answer: C
gives and , so and . The standard deviation is : that is why counting events has an uncertainty .Why people get this wrong: is the standard deviation, not the variance.
-
NATWarm-up
The expected value of the number shown by a fair six-sided die is ______.Answer and solution
Answer: 3.5 (±0.01)
Each face has probability : . The mean need not be a possible outcome. -
MCQWarm-up
For a Gaussian (normal) distribution, the probability that a value lies within one standard deviation of the mean is aboutAnswer and solution
Answer: B
. The "68-95-99.7 rule" gives about 68%, 95% and 99.7% within , and . -
NATWarm-up
A detector registers each incoming photon with probability 0.3. For 10 photons, the mean number registered is ______.Answer and solution
Answer: 3
The number registered is binomial with , , whose mean is . -
MCQWarm-up
A random variable is uniformly distributed on . Its variance isAnswer and solution
Answer: A
and . So .Why people get this wrong: is ; you must subtract the square of the mean.
-
MCQWarm-up
Two fair coins are tossed. The probability of getting at least one head isAnswer and solution
Answer: D
The complement, no heads at all, has probability . So . (Listing HH, HT, TH, TT gives the same.)Why people get this wrong: comes from treating "0, 1, 2 heads" as equally likely; they are not.
-
NATStandard
In 20 independent trials, each succeeding with probability 0.25, the variance of the number of successes is ______.Answer and solution
Answer: 3.75 (±0.01)
For the binomial distribution . -
MCQStandard
A Geiger counter records on average 2 counts per second from a weak source. The probability of no count in a given second isAnswer and solution
Answer: A
Counts in a fixed interval are Poisson with : . is and is the probability of at least one count. -
NATStandard
The waiting time between decays has the density with . The mean waiting time is ______ s.Answer and solution
Answer: 2 s (±0.01)
s. The standard deviation is also . -
MCQStandard
A particle takes steps of fixed length , each equally likely to the left or right, independently. Its root-mean-square distance from the start isAnswer and solution
Answer: B
Write with . Then and . So : the root of diffusion, . -
NATStandard
Two fair dice are thrown. The probability that the total is 7 is ______ (give a decimal).Answer and solution
Answer: 0.1667 (±0.002)
Of the 36 equally likely pairs, six give 7: (1,6), (2,5), (3,4), (4,3), (5,2), (6,1). . Seven is the most likely total. -
MCQStandard
If has variance and with constants , then the variance of isAnswer and solution
Answer: D
: the shift cancels. Squaring and averaging, .Why people get this wrong: Adding a constant moves the distribution without widening it.
-
MSQStandard
For two independent random variables and , which relations always hold?Answer and solution
Answer: A, B, C
The mean is linear for any variables. Independence makes the covariance zero, so and variances add. For a difference, : variances still add, they never subtract.Why people get this wrong: Errors add in quadrature for a difference as well as for a sum.
-
NATStandard
A detector records on average 400 counts in a fixed time window. The standard deviation of the number of counts from one window to the next is ______.Answer and solution
Answer: 20
Independent rare events give Poisson statistics, so , a 5% fractional spread. -
MCQStandard
A rare condition affects 1% of a population. A test detects it in 99% of those who have it, and falsely signals it in 1% of those who do not. A person tests positive. The probability that they have the condition isAnswer and solution
Answer: B
Bayes: . In 10 000 people: 99 true positives and 99 false positives.Why people get this wrong: Confusing with ignores how rare the condition is.
-
MCQStandard
The probability density on the whole real line is normalised when equalsAnswer and solution
Answer: D
, so . This is the Laplace (double-exponential) distribution, with variance 2. -
NATStandard
For a Gaussian distribution, the percentage of values that lie more than away from the mean (on either side) is ______ %.Answer and solution
Answer: 4.55 % (±0.1)
. The rest, , is 4.55%, split equally between the two tails. -
NATExam level
Each of 1000 independent nuclei decays in a given minute with probability 0.002. Using the Poisson approximation, the probability that none decays in that minute is ______.Answer and solution
Answer: 0.1353 (±0.002)
For large and small the binomial becomes Poisson with . . (The exact binomial value agrees.) -
MCQExam level
The Lorentzian (Cauchy) density describes a natural line shape. Which statement is correct?- Its mean is 0 and its variance is
- Its variance is infinite, and its mean is not defined because diverges
- Its variance is
- Its mean is
Answer and solution
Answer: B
For large , , so diverges logarithmically and diverges linearly. The width is described instead by the half width at half maximum, . Because the variance is infinite, the central limit theorem does not apply: the average of many Cauchy variables is still Cauchy with the same width.Why people get this wrong: Symmetry makes the principal value of the mean zero, but the mean as an expectation does not exist.
-
NATExam level
For a Gaussian with mean 0 and standard deviation , the fourth moment is ______.Answer and solution
Answer: 48
For a zero-mean Gaussian , so . The ratio is the Gaussian kurtosis. -
MCQExam level
The characteristic function of a Gaussian with mean and variance isAnswer and solution
Answer: C
Completing the square in gives . Its logarithm has only first and second cumulants ( and ); all higher cumulants of a Gaussian vanish. The Fourier transform of a Gaussian is a Gaussian. -
NATExam level
Two independent sources give Poisson counts with means 2 and 3 in a fixed interval. The probability that the total number of counts is exactly 1 is ______.Answer and solution
Answer: 0.0337 (±0.0005)
A sum of independent Poisson variables is Poisson with the summed mean, . . Check directly: . -
MCQExam level
is uniformly distributed on and . The probability density of is- uniform on
- for
- for
- for
Answer and solution
Answer: B
with : for . This is how computers turn uniform random numbers into exponential waiting times.Why people get this wrong: The Jacobian is essential: a density is not simply re-labelled.
-
NATExam level
For the Maxwell speed distribution of a gas, the ratio of the root-mean-square speed to the most probable speed is ______.Answer and solution
Answer: 1.2247 (±0.005)
(the peak of ) and (from ). The ratio is . The mean speed lies between them. -
MSQExam level
Which statements about the central limit theorem are correct?- The sum of independent, identically distributed variables with finite variance tends to a Gaussian for large
- The mean of such samples has standard deviation
- It applies to Cauchy-distributed variables
- The variance of the sum grows in proportion to
Answer and solution
Answer: A, B, D
Variances of independent variables add, so the sum has variance and the mean . The theorem needs a finite variance; the Cauchy distribution has none, and the mean of Cauchy variables has the same Cauchy distribution as one. -
NATExam level
A walker takes 10 unit steps, each to the left or right with probability 1/2. The probability that it ends at the starting point is ______.Answer and solution
Answer: 0.2461 (±0.002)
Ending at the origin needs exactly 5 steps right out of 10: . The Gaussian estimate (the factor 2 because only even positions are reachable) is close. -
MCQExam level
The lifetime of an unstable nucleus is exponentially distributed. A nucleus has already survived a time . The probability that it survives a further time isAnswer and solution
Answer: C
: the nucleus has no memory of its age. The exponential is the only continuous distribution with this property, which is why decay has a fixed half-life. -
NATExam level
takes the values with equal probability, and . The covariance is ______.Answer and solution
Answer: 0
and , so the covariance is 0. Yet is completely determined by : zero covariance does not mean independence. -
MCQExam level
The full width at half maximum of a Gaussian of standard deviation isAnswer and solution
Answer: B
Half maximum where : . That is the half width; the full width is twice it, .Why people get this wrong: is the half width at half maximum.
-
NATExam level
In 8 tosses of a fair coin, the probability of 6 or more heads is ______ (to three decimal places).Answer and solution
Answer: 0.145 (±0.002)
.Why people get this wrong: Using alone gives 0.109 and misses 7 and 8 heads.