Home / Courses / GATE Physics / Practice / Random variables and probability distributions

Random variables and probability distributions

Mathematical methods · 77 questions and templates for GATE Physics, in order from warm-up to top rank. Check each answer, then read the full solution.

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.

  1. MCQWarm-up

    For a Poisson distribution with mean λ, the variance is
    1. λ2
    2. λ
    3. λ
    4. 1/λ
    Answer and solution

    Answer: C

    P(n)=λne−λ/n! gives ⟨n⟩=λ and ⟨n(n−1)⟩=λ2, so ⟨n2⟩=λ2+λ and σ2=λ. The standard deviation is λ: that is why counting N events has an uncertainty N.

    Why people get this wrong: λ is the standard deviation, not the variance.

  2. 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 1/6: ⟨X⟩=16(1+2+3+4+5+6)=21/6=3.5. The mean need not be a possible outcome.
  3. MCQWarm-up

    For a Gaussian (normal) distribution, the probability that a value lies within one standard deviation of the mean is about
    1. 50%
    2. 68%
    3. 95%
    4. 99.7%
    Answer and solution

    Answer: B

    P(|x−μ|<σ)=erf(1/2)≈0.683. The "68-95-99.7 rule" gives about 68%, 95% and 99.7% within 1σ, 2σ and 3σ.
  4. 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 n=10, p=0.3, whose mean is np=3.
  5. MCQWarm-up

    A random variable is uniformly distributed on [0,a]. Its variance is
    1. a2/12
    2. a2/4
    3. a/2
    4. a2/3
    Answer and solution

    Answer: A

    ⟨x⟩=a/2 and ⟨x2⟩=1a∫0ax2dx=a2/3. So σ2=a2/3−a2/4=a2/12.

    Why people get this wrong: a2/3 is ⟨x2⟩; you must subtract the square of the mean.

  6. MCQWarm-up

    Two fair coins are tossed. The probability of getting at least one head is
    1. 1/2
    2. 1/4
    3. 2/3
    4. 3/4
    Answer and solution

    Answer: D

    The complement, no heads at all, has probability 12⋅12=14. So P(at least one head)=1−14=34. (Listing HH, HT, TH, TT gives the same.)

    Why people get this wrong: 2/3 comes from treating "0, 1, 2 heads" as equally likely; they are not.

  7. MCQWarm-up

    A variable is uniformly distributed on the interval [2,8]. Its mean is
    1. 4
    2. 5
    3. 6
    4. 3
    Answer and solution

    Answer: B

    A uniform density is flat, so the mean is the midpoint: (2+8)/2=5.

    Why people get this wrong: 6/2=3 is half the width, not the centre.

  8. 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 σ2=np(1−p)=20×0.25×0.75=3.75.
  9. MCQStandard

    A Geiger counter records on average 2 counts per second from a weak source. The probability of no count in a given second is
    1. e−2
    2. 2e−2
    3. 1−e−2
    4. e−1/2
    Answer and solution

    Answer: A

    Counts in a fixed interval are Poisson with λ=2: P(0)=λ0e−λ/0!=e−2≈0.135. 2e−2 is P(1) and 1−e−2 is the probability of at least one count.
  10. NATStandard

    The waiting time t between decays has the density p(t)=λe−λt with λ=0.5 s−1. The mean waiting time is ______ s.
    Answer and solution

    Answer: 2 s (±0.01)

    ⟨t⟩=∫0∞tλe−λtdt=1/λ=2 s. The standard deviation is also 1/λ.
  11. MCQStandard

    A particle takes N steps of fixed length l, each equally likely to the left or right, independently. Its root-mean-square distance from the start is
    1. Nl
    2. lN
    3. lN2
    4. l/N
    Answer and solution

    Answer: B

    Write x=∑isi with si=±l. Then ⟨x⟩=0 and ⟨x2⟩=∑i⟨si2⟩+∑i≠j⟨si⟩⟨sj⟩=Nl2. So xrms=lN: the root of diffusion, ⟨x2⟩∝t.
  12. 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). P=6/36=1/6≈0.167. Seven is the most likely total.
  13. MCQStandard

    If X has variance σ2 and Y=aX+b with constants a,b, then the variance of Y is
    1. aσ2+b
    2. a2σ2+b2
    3. |a|σ2
    4. a2σ2
    Answer and solution

    Answer: D

    Y−⟨Y⟩=a(X−⟨X⟩): the shift b cancels. Squaring and averaging, VarY=a2σ2.

    Why people get this wrong: Adding a constant moves the distribution without widening it.

  14. MSQStandard

    For two independent random variables X and Y, which relations always hold?
    1. ⟨X+Y⟩=⟨X⟩+⟨Y⟩
    2. Var(X+Y)=VarX+VarY
    3. ⟨XY⟩=⟨X⟩⟨Y⟩
    4. Var(X−Y)=VarX−VarY
    Answer and solution

    Answer: A, B, C

    The mean is linear for any variables. Independence makes the covariance zero, so ⟨XY⟩=⟨X⟩⟨Y⟩ and variances add. For a difference, Var(X−Y)=VarX+(−1)2VarY: variances still add, they never subtract.

    Why people get this wrong: Errors add in quadrature for a difference as well as for a sum.

  15. 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 σ=⟨N⟩=400=20, a 5% fractional spread.
  16. 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 is
    1. 99%
    2. 50%
    3. 1%
    4. 90%
    Answer and solution

    Answer: B

    Bayes: P(C|+)=P(+|C)P(C)P(+|C)P(C)+P(+|C‾)P(C‾)=0.99×0.010.99×0.01+0.01×0.99=12. In 10 000 people: 99 true positives and 99 false positives.

    Why people get this wrong: Confusing P(+|C)=99% with P(C|+) ignores how rare the condition is.

  17. MCQStandard

    The probability density p(x)=Ce−|x| on the whole real line is normalised when C equals
    1. 1
    2. 2
    3. 1/e
    4. 1/2
    Answer and solution

    Answer: D

    ∫−∞∞e−|x|dx=2∫0∞e−xdx=2, so C=1/2. This is the Laplace (double-exponential) distribution, with variance 2.
  18. NATStandard

    For a Gaussian distribution, the percentage of values that lie more than 2σ away from the mean (on either side) is ______ %.
    Answer and solution

    Answer: 4.55 % (±0.1)

    P(|x−μ|<2σ)=erf(2/2)=erf(1.414)≈0.9545. The rest, 1−0.9545=0.0455, is 4.55%, split equally between the two tails.
  19. MCQStandard

    For a Poisson distribution with mean λ=3.5, the most probable count is
    1. 3
    2. 4
    3. 3.5
    4. 3 and 4 equally
    Answer and solution

    Answer: A

    P(n)P(n−1)=λn. Going from 3 to 4 the ratio is 3.5/4<1, so P(4)<P(3); going from 2 to 3 it is 3.5/3>1. The peak is at n=3, the integer part of λ. (Two equal peaks happen only for integer λ.)

    Why people get this wrong: A count must be a whole number, so 3.5 cannot be the mode even though it is the mean.

  20. NATStandard

    A fair die is rolled 6 times. The probability that each face appears exactly once is ______ (four decimal places).
    Answer and solution

    Answer: 0.0154 (±0.0002)

    Favourable sequences: the six faces in any order, 6!=720. All sequences: 66=46656. Probability 72046656≈0.0154.

    Why people get this wrong: Because there are six faces and six rolls, it feels likely; it happens only about once in 65 tries.

  21. 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 n and small p the binomial becomes Poisson with λ=np=2. P(0)=e−2≈0.135. (The exact binomial value 0.9981000=e1000ln⁡0.998≈e−2.002≈0.135 agrees.)
  22. NATExam level

    For a Gaussian with mean 0 and standard deviation σ=2, the fourth moment ⟨x4⟩ is ______.
    Answer and solution

    Answer: 48

    For a zero-mean Gaussian ⟨x2n⟩=(2n−1)!!σ2n, so ⟨x4⟩=3σ4=3×16=48. The ratio ⟨x4⟩/⟨x2⟩2=3 is the Gaussian kurtosis.
  23. 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, λ=5. P(1)=5e−5≈0.0337. Check directly: P=P1(1)P2(0)+P1(0)P2(1)=2e−2e−3+e−23e−3=5e−5.
  24. MCQExam level

    X is uniformly distributed on (0,1) and Y=−ln⁡X. The probability density of Y is
    1. uniform on (0,∞)
    2. e−y for y>0
    3. ye−y for y>0
    4. 1/y for y>0
    Answer and solution

    Answer: B

    pY(y)=pX(x)|dxdy| with x=e−y: pY=1⋅e−y for y>0. This is how computers turn uniform random numbers into exponential waiting times.

    Why people get this wrong: The Jacobian |dx/dy| is essential: a density is not simply re-labelled.

  25. NATExam level

    For the Maxwell speed distribution of a gas, the ratio vrms/vp of the root-mean-square speed to the most probable speed is ______.
    Answer and solution

    Answer: 1.2247 (±0.005)

    vp=2kT/m (the peak of v2e−mv2/2kT) and vrms=3kT/m (from 12m⟨v2⟩=32kT). The ratio is 3/2≈1.225. The mean speed 8kT/πm lies between them.
  26. MSQExam level

    Which statements about the central limit theorem are correct?
    1. The sum of N independent, identically distributed variables with finite variance tends to a Gaussian for large N
    2. The mean of N such samples has standard deviation σ/N
    3. It applies to Cauchy-distributed variables
    4. The variance of the sum grows in proportion to N
    Answer and solution

    Answer: A, B, D

    Variances of independent variables add, so the sum has variance Nσ2 and the mean σ2/N. The theorem needs a finite variance; the Cauchy distribution has none, and the mean of N Cauchy variables has the same Cauchy distribution as one.
  27. 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: P=(105)/210=252/1024≈0.246. The Gaussian estimate 2/2πN=0.252 (the factor 2 because only even positions are reachable) is close.
  28. MCQExam level

    The lifetime T of an unstable nucleus is exponentially distributed. A nucleus has already survived a time s. The probability that it survives a further time t is
    1. e−λ(s+t)
    2. e−λs
    3. e−λt
    4. 1−e−λt
    Answer and solution

    Answer: C

    P(T>s+t|T>s)=e−λ(s+t)e−λs=e−λt: 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.
  29. NATExam level

    X takes the values −1,0,1 with equal probability, and Y=X2. The covariance ⟨XY⟩−⟨X⟩⟨Y⟩ is ______.
    Answer and solution

    Answer: 0

    ⟨XY⟩=⟨X3⟩=13(−1+0+1)=0 and ⟨X⟩=0, so the covariance is 0. Yet Y is completely determined by X: zero covariance does not mean independence.
  30. MCQExam level

    The full width at half maximum of a Gaussian of standard deviation σ is
    1. 2σ
    2. 22ln⁡2σ≈2.355σ
    3. 2ln⁡2σ≈1.177σ
    4. 2σ
    Answer and solution

    Answer: B

    Half maximum where e−x2/2σ2=12: x=σ2ln⁡2≈1.177σ. That is the half width; the full width is twice it, 2.355σ.

    Why people get this wrong: 1.177σ is the half width at half maximum.

← All topics · Timed mock test →