Term 1 · Module 4 of 4

Continuous Probability Distribution

Business Statistics for Entrepreneurs I

Practise this module in Claude, ChatGPT or Gemini

Set it up free
Mind map

Continuous Random Variables: From Point Probabilities to Areas

A continuous random variable measures a quantity that can take any value in a range, including fractional values: waiting time, fill volume, temperature, or distance between arrivals. A discrete variable counts distinct outcomes, such as the number of defects. This difference changes how probability is calculated: a continuous model assigns probability to intervals through area, not to individual points.

Let XX denote the random variable and xx a possible value. Its probability density function (PDF) f(x)f(x) is nonnegative and has total area 1. The height f(x)f(x) is a density, not the probability of exactly xx.

f(x)≥0,∫range of Xf(x) dx=1f(x)\geq 0,\qquad \int_{\text{range of }X} f(x)\,dx=1

For an interval, probability is the area under the density curve:

Pr⁡(a≤X≤b)=∫abf(x) dx\Pr(a\leq X\leq b)=\int_a^b f(x)\,dx

A single point has zero width and therefore zero area: Pr⁡(X=a)=0\Pr(X=a)=0. For a continuous variable, including or excluding an interval endpoint does not change its probability. Integration plays the role that summation plays for a discrete probability mass function; tables or software can provide the result without integrating from first principles.

The cumulative distribution function (CDF) accumulates probability up to a value, while the tail probability captures values above it:

F(x)=Pr⁡(X≤x)=∫−∞xf(t) dt,F‾(x)=Pr⁡(X>x)=1−F(x)F(x)=\Pr(X\leq x)=\int_{-\infty}^{x}f(t)\,dt,\qquad \overline F(x)=\Pr(X>x)=1-F(x)

The lower limit is understood to be the minimum possible value when the variable has a restricted range. The expected value is the probability-weighted center and the variance measures squared spread around it:

E[X]=∫xf(x) dx,Var⁡(X)=∫(x−E[X])2f(x) dx\mathbb E[X]=\int x f(x)\,dx,\qquad \operatorname{Var}(X)=\int (x-\mathbb E[X])^2 f(x)\,dx

Question Probability expression
At most bb F(b)F(b)
Greater than aa 1−F(a)=F‾(a)1-F(a)=\overline F(a)
Between aa and bb F(b)−F(a)F(b)-F(a)
Exactly aa 00

Key takeaways

  • A continuous variable can take fractional values; probability concerns intervals.
  • A PDF is a density; area under it gives probability.
  • The full area is 1, and any single point has probability 0.
  • CDFs, tails, expectation, and variance extend familiar discrete ideas using integration.

Uniform Distribution

A uniform random variable assigns equal density throughout a finite interval [L,U][L,U]. Since the density is constant, probability is proportional to interval length: an interval twice as long is twice as likely.

f(x)={1U−L,L≤x≤U0,otherwisef(x)=\begin{cases}\dfrac{1}{U-L},&L\leq x\leq U\\0,&\text{otherwise}\end{cases}

For L≤x≤UL\leq x\leq U, the CDF and upper tail are:

F(x)=x−LU−L,F‾(x)=U−xU−LF(x)=\frac{x-L}{U-L},\qquad \overline F(x)=\frac{U-x}{U-L}

Therefore, for L≤a≤b≤UL\leq a\leq b\leq U:

Pr⁡(a≤X≤b)=b−aU−L,E[X]=L+U2,Var⁡(X)=(U−L)212\Pr(a\leq X\leq b)=\frac{b-a}{U-L},\qquad \mathbb E[X]=\frac{L+U}{2},\qquad \operatorname{Var}(X)=\frac{(U-L)^2}{12}

Uniform Distribution

Driving time example

A trip takes between 100 and 140 minutes and is modeled uniformly. Thus L=100L=100, U=140U=140, and the mean is (100+140)/2=120(100+140)/2=120 minutes. The density is 1/(140−100)=0.0251/(140-100)=0.025 per minute.

  • At most 130 minutes:

F(130)=130−100140−100=3040=0.75F(130)=\frac{130-100}{140-100}=\frac{30}{40}=0.75

  • More than 105 minutes:

F‾(105)=140−105140−100=3540=0.875\overline F(105)=\frac{140-105}{140-100}=\frac{35}{40}=0.875

  • Exactly 120 minutes: Pr⁡(X=120)=0\Pr(X=120)=0, despite the positive PDF height of 0.0250.025. The density is not a point probability.

Phone battery example

Battery time to recharge ranges uniformly from 8 to 12 hours. Here L=8L=8, U=12U=12, and mean time is (8+12)/2=10(8+12)/2=10 hours.

  • Recharge needed by 9 hours:

Pr⁡(X≤9)=9−812−8=14=0.25\Pr(X\leq9)=\frac{9-8}{12-8}=\frac14=0.25

  • Battery lasts at least 11 hours:

Pr⁡(X≥11)=12−1112−8=14=0.25\Pr(X\geq11)=\frac{12-11}{12-8}=\frac14=0.25

  • 80th percentile: solve (x−8)/(12−8)=0.8(x-8)/(12-8)=0.8 to get x=8+3.2=11.2x=8+3.2=11.2 hours. Thus 80% of battery durations are at most 11.2 hours.

The discrete counterpart is a fair die: its six distinct outcomes each have probability 1/61/6. The continuous uniform model instead spreads probability evenly over an interval.

Key takeaways

  • Uniform probability is interval length divided by total range.
  • The mean is the midpoint; variance depends on squared interval width.
  • A percentile is found by setting the CDF equal to its probability and solving for xx.
  • A density value such as 0.0250.025 is not the probability of one exact time.

Exponential Distribution: Time Between Random Events

When events occur randomly at a constant average rate λ\lambda, the exponential distribution often models the waiting time between consecutive events. It measures time or distance between events, not the event count. Its support is nonnegative, and its mean is 1/λ1/\lambda.

f(x)=λe−λx,x≥0;F(x)=1−e−λx;F‾(x)=e−λxf(x)=\lambda e^{-\lambda x},\quad x\geq0;\qquad F(x)=1-e^{-\lambda x};\qquad \overline F(x)=e^{-\lambda x}

For 0≤a<b0\leq a<b:

Pr⁡(a<X≤b)=e−λa−e−λb,E[X]=1λ,Var⁡(X)=1λ2,SD⁡(X)=1λ\Pr(a<X\leq b)=e^{-\lambda a}-e^{-\lambda b},\qquad \mathbb E[X]=\frac1\lambda,\qquad \operatorname{Var}(X)=\frac1{\lambda^2},\qquad \operatorname{SD}(X)=\frac1\lambda

The mean and standard deviation are equal. The density is right-skewed. Common applications include manual task duration, call duration, equipment lifetime until repair, and waiting in a queue, when the constant-rate model is appropriate.

Exponential Distribution: Time Between Random Events

Customer-service wait example

Average wait is 2 minutes, so 1/λ=21/\lambda=2 and λ=0.5\lambda=0.5 per minute.

  • Less than 1 minute:

Pr⁡(X≤1)=1−e−0.5(1)=1−e−0.5≈0.393\Pr(X\leq1)=1-e^{-0.5(1)}=1-e^{-0.5}\approx0.393

  • More than 3 minutes:

Pr⁡(X>3)=e−0.5(3)=e−1.5≈0.223\Pr(X>3)=e^{-0.5(3)}=e^{-1.5}\approx0.223

  • Between 1 and 3 minutes:

Pr⁡(1<X<3)=e−0.5(1)−e−0.5(3)≈0.607−0.223=0.383\Pr(1<X<3)=e^{-0.5(1)}-e^{-0.5(3)}\approx0.607-0.223=0.383

  • 90th percentile: solve 1−e−0.5x=0.91-e^{-0.5x}=0.9, so e−0.5x=0.1e^{-0.5x}=0.1 and x=−ln⁡(0.1)/0.5≈4.605x=-\ln(0.1)/0.5\approx4.605 minutes.

Memoryless property and bus waiting

The exponential distribution is memoryless: if an event has not happened by time t0t_0, the additional wait has the same exponential distribution as the original wait. The elapsed wait does not change the remaining-wait distribution.

Pr⁡(X>t0+y∣X>t0)=e−λ(t0+y)e−λt0=e−λy=Pr⁡(X>y)\Pr(X>t_0+y\mid X>t_0)=\frac{e^{-\lambda(t_0+y)}}{e^{-\lambda t_0}}=e^{-\lambda y}=\Pr(X>y)

A bus wait averages 20 minutes, hence λ=1/20=0.05\lambda=1/20=0.05 per minute.

  • Wait 15 minutes or less: 1−e−0.05(15)=1−e−0.75≈0.5281-e^{-0.05(15)}=1-e^{-0.75}\approx0.528.
  • Wait more than 30 minutes: e−0.05(30)=e−1.5≈0.223e^{-0.05(30)}=e^{-1.5}\approx0.223.
  • Wait between 15 and 30 minutes: e−0.75−e−1.5≈0.472−0.223=0.249e^{-0.75}-e^{-1.5}\approx0.472-0.223=0.249.
  • Having already waited 10 minutes, wait at least 10 more: Pr⁡(X>20∣X>10)=e−0.05(10)≈0.607\Pr(X>20\mid X>10)=e^{-0.05(10)}\approx0.607.
  • Having already waited 20 minutes, wait at least 10 more: Pr⁡(X>30∣X>20)=e−0.05(10)≈0.607\Pr(X>30\mid X>20)=e^{-0.05(10)}\approx0.607 again.

Connection to the Poisson process

A Poisson process counts random events over time at a constant rate. If gaps between successive events are independent exponential random variables with rate λ\lambda, the count during a time interval of length tt has a Poisson distribution with mean Λ=λt\Lambda=\lambda t. When gaps are deterministic, the count is deterministic; when gaps are exponential, the count is random.

For highway defects averaging 3 per kilometre, the mean distance between defects is 1/31/3 km, and the exponential rate is 3 per kilometre.

  • At least 0.5 km with no defect: Pr⁡(X>0.5)=e−3(0.5)=e−1.5≈0.223\Pr(X>0.5)=e^{-3(0.5)}=e^{-1.5}\approx0.223.
  • At least 2 km with no defect: Pr⁡(X>2)=e−3(2)=e−6≈0.0025\Pr(X>2)=e^{-3(2)}=e^{-6}\approx0.0025 (about 0.002 using the displayed table precision).

Oil-change time and counts

An oil change takes an exponentially distributed time with mean 20 minutes, so λ=0.05\lambda=0.05 per minute.

  • Finish within 20 minutes: F(20)=1−e−0.05(20)=1−e−1≈0.632F(20)=1-e^{-0.05(20)}=1-e^{-1}\approx0.632. This is the probability relevant to a 20-minute guarantee.
  • Finish between 15 and 30 minutes: e−0.75−e−1.5≈0.249e^{-0.75}-e^{-1.5}\approx0.249.

The corresponding number completed in a fixed time follows a Poisson model, with mean equal to the rate times elapsed time. In a 4-hour (240-minute) shift, Λ=0.05(240)=12\Lambda=0.05(240)=12. Thus more than 15 means at least 16:

Pr⁡(Y>15)=1−Pr⁡(Y≤15),Y∼Poisson⁡(12)≈0.228\Pr(Y>15)=1-\Pr(Y\leq15),\quad Y\sim\operatorname{Poisson}(12)\approx0.228

In an 8-hour (480-minute) day, Λ=0.05(480)=24\Lambda=0.05(480)=24, and the probability of more than 30 (at least 31) completions is approximately 0.1320.132 as reported by the Poisson table.

Key takeaways

  • Exponential models waiting time; Poisson models event count over a fixed interval.
  • Convert mean wait to rate using λ=1/mean\lambda=1/\text{mean}, keeping units consistent.
  • Memorylessness means the additional wait distribution is unchanged by elapsed time.
  • For a Poisson process, expected count over duration tt is λt\lambda t.

Normal Distribution and Standard Normal Calculations

The normal distribution, also called the Gaussian or bell-shaped distribution, models many measurements and is central to statistical inference. Examples include heights, weights, test scores, scientific measurements and rainfall. A normal variable is described by its mean μ\mu and standard deviation σ\sigma:

f(x)=1σ2πexp⁡ ⁣(−(x−μ)22σ2),−∞<x<∞f(x)=\frac{1}{\sigma\sqrt{2\pi}}\exp\!\left(-\frac{(x-\mu)^2}{2\sigma^2}\right),\qquad -\infty<x<\infty

It is symmetric around μ\mu, where mean, median and mode coincide. Half the area lies on each side; skewness is zero. The tails extend indefinitely, approaching but never touching the horizontal axis. A larger σ\sigma makes the curve wider and flatter; a smaller σ\sigma makes it narrower and taller. The empirical coverage rule is approximately:

Range around the mean Probability
μ±σ\mu\pm\sigma 0.683
μ±2σ\mu\pm2\sigma 0.954
μ±3σ\mu\pm3\sigma 0.997

The standard normal variable ZZ has mean 0 and standard deviation 1. Its density is ϕ(z)=12πe−z2/2\phi(z)=\frac{1}{\sqrt{2\pi}}e^{-z^2/2}; its CDF is often written Φ(z)\Phi(z). Any normal variable can be standardized:

Z=X−μσ,X=μ+ZσZ=\frac{X-\mu}{\sigma},\qquad X=\mu+Z\sigma

The resulting zz-score counts standard deviations above (positive) or below (negative) the mean. Tables/software provide CDFs, tails, interval probabilities, and inverse percentiles. For example, if X∼N(10,22)X\sim N(10,2^2), then X=10X=10 maps to z=0z=0, and X=14X=14 maps to z=2z=2; hence Pr⁡(10<X<14)=Pr⁡(0<Z<2)≈0.4772\Pr(10<X<14)=\Pr(0<Z<2)\approx0.4772.

For a normal model with mean 10 and standard deviation 3, the CDF at 13 is 0.841, the upper tail at 13 is 0.159, and the interval from 7 to 13 has probability 0.683. In Excel, NORM.DIST(x, mean, standard_dev, TRUE) returns the CDF and FALSE returns density; NORM.INV(probability, mean, standard_dev) returns the value at a cumulative probability. Inverse upper-tail questions use left cumulative probability 1−p1-p.

Normal Distribution and Standard Normal Calculations

Standard-normal probability practice

CDF and tail results below are approximate table values. Since ZZ is continuous, no probability is double-counted at a shared endpoint.

Quantity Result
Pr⁡(Z≤−1)\Pr(Z\leq-1) 0.159
Pr⁡(Z≥−1)\Pr(Z\geq-1) 0.841
Pr⁡(Z≥−1.5)\Pr(Z\geq-1.5) 0.933
Pr⁡(Z≥−2.5)\Pr(Z\geq-2.5) 0.994
Pr⁡(−3<Z≤0)\Pr(-3<Z\leq0) 0.499
Pr⁡(0<Z<0.8)\Pr(0<Z<0.8) 0.288
Pr⁡(−1.5<Z<0)\Pr(-1.5<Z<0) 0.433
Pr⁡(Z≥0.4)\Pr(Z\geq0.4) 0.345
Pr⁡(Z≥−0.2)\Pr(Z\geq-0.2) 0.579
Pr⁡(Z≤1.2)\Pr(Z\leq1.2) 0.885
Pr⁡(Z≤−0.7)\Pr(Z\leq-0.7) 0.242

Inverse examples: left area 0.21 corresponds to z≈−0.806z\approx-0.806; central area 0.90 is between about −1.64-1.64 and 1.641.64; central area 0.20 is between about −0.25-0.25 and 0.250.25; left area 0.955 corresponds to z≈1.70z\approx1.70 (the transcript gives 2.58, which does not match this cumulative probability); and right area 0.692 corresponds to z≈−0.502z\approx-0.502.

A two-sided critical value zα/2z_{\alpha/2} leaves area α/2\alpha/2 in each tail, so the central area between −zα/2-z_{\alpha/2} and zα/2z_{\alpha/2} is 1−α1-\alpha.

Central area 1−α1-\alpha Tail area α/2\alpha/2 zα/2z_{\alpha/2} (approx.)
0.68 0.16 1.00
0.80 0.10 1.28
0.90 0.05 1.64
0.95 0.025 1.96
0.99 0.005 2.58

Sketching the bell curve and shading the requested area helps distinguish left tail, right tail and central intervals before consulting a table.

Investment-return example

Annual returns have a normal model with μ=12%\mu=12\% and σ=3\sigma=3 percentage points.

  • Unsatisfactory, below 5%: z=(5−12)/3=−2.33z=(5-12)/3=-2.33, giving Pr⁡(X<5%)≈0.01\Pr(X<5\%)\approx0.01.
  • Excellent, above 10%: z=(10−12)/3=−0.67z=(10-12)/3=-0.67, so the right tail is approximately 0.748.
  • Moderate, between 5% and 10%: Pr⁡(5%<X<10%)≈0.243\Pr(5\%<X<10\%)\approx0.243.
  • 90th percentile: x=12+1.281(3)≈15.8%x=12+1.281(3)\approx15.8\%. A return above this is in the top 10%.
  • A 90% upper-tail probability corresponds to the 10th percentile, about 8.1%. A central 90% interval is about 7% to 17%.

Screen-time example

Daily screen time has μ=8.4\mu=8.4 hours and σ=2.5\sigma=2.5 hours.

  • More than 4 hours: z=(4−8.4)/2.5=−1.76z=(4-8.4)/2.5=-1.76, so Pr⁡(X>4)≈0.96\Pr(X>4)\approx0.96.
  • Between 6 and 12 hours: approximately 0.757.
  • Top 20% threshold is the 80th percentile: x≈10.5x\approx10.5 hours; more time places a child in the top fifth.
  • Bottom 20% threshold is the 20th percentile: x≈6.3x\approx6.3 hours. Equivalently, 80% exceed about 6.3 hours.
  • A central 80% interval is about 5.2 to 11.6 hours.

Exam-completion times

Completion time has μ=150\mu=150 minutes, σ=20\sigma=20 minutes, for 60 students.

  • At most 2 hours (120 minutes): z=(120−150)/20=−1.5z=(120-150)/20=-1.5; probability 0.067. Expected number: 60(0.067)≈460(0.067)\approx4 students.
  • Between 2 and 3 hours (120 to 180 minutes): probability 0.866; expected number 60(0.866)≈5260(0.866)\approx52.
  • At least 3 hours (180 minutes): probability 0.067; expected number about 4. The equal tails follow from symmetry. Convert hours to minutes before using the model.

Choosing between filling machines

Packet weights are normal. Compare their probabilities against the target range rather than judging by mean alone.

Machine Mean, SD (kg) Below 1 kg Between 1 and 1.05 kg Central 90% interval (kg)
1 1.01, 0.02 0.309 0.669 0.97 to 1.04
2 1.03, 0.04 0.227 0.465 0.96 to 1.09

For machine 1, the 90% interval is centered on 1.01 and has half-width approximately 1.645(0.02)=0.0331.645(0.02)=0.033, yielding about 0.98 to 1.04 by direct calculation; the transcript/template reports 0.97 to 1.04. For machine 2, the reported interval is 0.96 to 1.09. Neither machine puts 90% of packets between 0.99 and 1.01 kg. Machine 2 has a lower underweight probability in this comparison, but fewer packets in the 1 to 1.05 kg band and greater spread; the choice depends on the quality criterion.

Glass-sheet thickness

Thickness is normal with μ=3\mu=3 mm and σ=0.12\sigma=0.12 mm.

  • Below 2.9 mm: z=(2.9−3)/0.12≈−0.83z=(2.9-3)/0.12\approx-0.83, so probability is 0.202.
  • Above 3.2 mm: z=(3.2−3)/0.12≈1.67z=(3.2-3)/0.12\approx1.67, so upper-tail probability is 0.048.
  • Central 95% interval: approximately 3±1.96(0.12)=3±0.2353\pm1.96(0.12)=3\pm0.235, or 2.76 to 3.24 mm.

Key takeaways

  • A normal model is set by μ\mu and σ\sigma; symmetry makes the two halves equal.
  • Standardization expresses distance from the mean in standard deviations.
  • Left area, right tail, interval, and inverse-percentile questions are distinct; sketch the requested region.
  • Keep units consistent, especially for time and percentiles.

Normal Approximation to the Binomial Distribution

The binomial distribution is discrete, while the normal is continuous. When the binomial shape is sufficiently bell-like, the normal can approximate its probabilities, avoiding cumbersome factorial calculations. A common rule in this module is np≥5np\geq5 and nq≥5nq\geq5, where q=1−pq=1-p. Match the binomial mean and variance:

X∼Binomial⁡(n,p),μ=np,σ2=npq,Y∼N(np,npq)X\sim\operatorname{Binomial}(n,p),\qquad \mu=np,\qquad \sigma^2=npq,\qquad Y\sim N(np,npq)

Because a continuous variable has zero probability at a point, use a continuity correction to include the discrete integer outcomes in a half-unit-wide interval:

Binomial event Normal approximation event
X≤kX\leq k Y≤k+0.5Y\leq k+0.5
X≥kX\geq k Y≥k−0.5Y\geq k-0.5
a≤X≤ba\leq X\leq b a−0.5≤Y≤b+0.5a-0.5\leq Y\leq b+0.5
X=kX=k k−0.5≤Y≤k+0.5k-0.5\leq Y\leq k+0.5

Normal Approximation to the Binomial Distribution

Sixteen fair-coin tosses

For n=16n=16, p=q=0.5p=q=0.5, so μ=16(0.5)=8\mu=16(0.5)=8, variance 16(0.5)(0.5)=416(0.5)(0.5)=4, and σ=2\sigma=2.

  • At most 5 successes, exactly: Pr⁡(X≤5)=0.105\Pr(X\leq5)=0.105.
  • With continuity correction, Pr⁡(Y≤5.5)\Pr(Y\leq5.5) for Y∼N(8,22)Y\sim N(8,2^2) is approximately 0.1056.
  • Between 8 and 11 successes inclusive, exact binomial probability is 0.5598. Approximate with Pr⁡(7.5≤Y≤11.5)≈0.5586\Pr(7.5\leq Y\leq11.5)\approx0.5586.

Invoice errors

Among n=100n=100 invoices, error probability is p=0.1p=0.1; q=0.9q=0.9. Mean np=10np=10, variance npq=9npq=9, and standard deviation 33. Conditions hold: np=10≥5np=10\geq5 and nq=90≥5nq=90\geq5.

  • Exactly 12 errors: binomial probability is about 0.09. Approximate by the normal area from 11.5 to 12.5:

Pr⁡(11.5≤Y≤12.5),Y∼N(10,32)≈0.1052\Pr(11.5\leq Y\leq12.5),\quad Y\sim N(10,3^2)\approx0.1052

  • At most 13 errors: correction gives Pr⁡(Y≤13.5)≈0.879\Pr(Y\leq13.5)\approx0.879; exact binomial value is about 0.876.

These are approximations, and accuracy generally improves when both npnp and nqnq are sufficiently large.

Startup funding example

Of 200 seed-funded startups, 20% receive the next funding level. Let XX count successful transitions: n=200n=200, p=0.2p=0.2, q=0.8q=0.8.

μ=np=200(0.2)=40,σ2=npq=200(0.2)(0.8)=32,σ=32≈5.66\mu=np=200(0.2)=40,\qquad \sigma^2=npq=200(0.2)(0.8)=32,\qquad \sigma=\sqrt{32}\approx5.66

Both conditions hold: np=40np=40 and nq=160nq=160 exceed 5.

  • At least 45: apply lower-bound correction, Pr⁡(X≥45)≈Pr⁡(Y≥44.5)\Pr(X\geq45)\approx\Pr(Y\geq44.5). Standardize: z=(44.5−40)/5.66≈0.795z=(44.5-40)/5.66\approx0.795. The upper tail is about 0.2148, or 21%.
  • At most 50: upper-bound correction gives Pr⁡(X≤50)≈Pr⁡(Y≤50.5)\Pr(X\leq50)\approx\Pr(Y\leq50.5). Here z=(50.5−40)/5.66≈1.86z=(50.5-40)/5.66\approx1.86, giving about 0.9686, or 97%.

Normal approximations are useful computationally, but remain estimates of the exact binomial probabilities. The Poisson distribution also has mean and variance both equal to its rate parameter; the closing passage on approximating Poisson with binomial is internally inconsistent, so no additional approximation rule is inferred here.

Key takeaways

  • Approximate a binomial with N(np,npq)N(np,npq) when both expected successes and failures are sufficiently large.
  • Preserve the discrete event using a half-unit continuity correction.
  • Match the mean and variance before standardizing.
  • State results as approximations and retain exact binomial values when available.

Additional Worked Examples

Battery life under two usage patterns

Talk-time battery life averages 20 hours; internet-use battery life averages 7 hours. Model each as a separate exponential variable, XX and YY, with different rates.

  • Talk time: λX=1/20\lambda_X=1/20. No more than 15 hours:

Pr⁡(X≤15)=1−e−(1/20)(15)=1−e−0.75≈0.5276\Pr(X\leq15)=1-e^{-(1/20)(15)}=1-e^{-0.75}\approx0.5276

  • Talk time: more than 20 hours:

Pr⁡(X>20)=e−(1/20)(20)=e−1≈0.3678\Pr(X>20)=e^{-(1/20)(20)}=e^{-1}\approx0.3678

  • Internet use: λY=1/7\lambda_Y=1/7. No more than 5 hours:

Pr⁡(Y≤5)=1−e−(1/7)(5)=1−e−5/7≈0.5105\Pr(Y\leq5)=1-e^{-(1/7)(5)}=1-e^{-5/7}\approx0.5105

Identify the correct usage-specific random variable and mean before calculating; the exponential rate is the reciprocal of the mean.

Website visits: counts and gaps

A website receives an average of 6 visits per minute. The mean gap is 60/6=1060/6=10 seconds, so interarrival time is exponential with mean 10 seconds and rate 1/101/10 per second. Counts are Poisson with mean rate times duration.

  1. Mean time between visits: 60/6=1060/6=10 seconds.
  2. No visits in one minute: for X∼Poisson⁡(6)X\sim\operatorname{Poisson}(6),

Pr⁡(X=0)=e−6≈0.0025\Pr(X=0)=e^{-6}\approx0.0025

  1. At most 10 visits in two minutes: Y∼Poisson⁡(6×2=12)Y\sim\operatorname{Poisson}(6\times2=12), so the table gives Pr⁡(Y≤10)=0.3742\Pr(Y\leq10)=0.3742.
  2. Gap at most 15 seconds: for ZZ exponential with λ=0.1\lambda=0.1 per second,

Pr⁡(Z≤15)=1−e−0.1(15)=1−e−1.5≈0.777\Pr(Z\leq15)=1-e^{-0.1(15)}=1-e^{-1.5}\approx0.777

  1. Gap more than 20 seconds:

Pr⁡(Z>20)=e−0.1(20)=e−2≈0.135\Pr(Z>20)=e^{-0.1(20)}=e^{-2}\approx0.135

This example switches between the Poisson count and exponential gap while keeping the same event rate and consistent time units.

Investment-fund returns

Three-year returns have mean 20% and standard deviation 4 percentage points. Let X∼N(20,42)X\sim N(20,4^2).

  • Return above 25%: z=(25−20)/4=1.25z=(25-20)/4=1.25. The CDF at 1.25 is 0.894, so

Pr⁡(X>25)=Pr⁡(Z>1.25)=1−0.894=0.106\Pr(X>25)=\Pr(Z>1.25)=1-0.894=0.106

  • Return at most 10%: z=(10−20)/4=−2.5z=(10-20)/4=-2.5, so Pr⁡(X≤10)=Pr⁡(Z≤−2.5)=0.0062\Pr(X\leq10)=\Pr(Z\leq-2.5)=0.0062.
  • Threshold for top 10%: left cumulative probability is 0.90, corresponding to z=1.281z=1.281. Convert back:

x=20+1.281(4)=25.124%≈25.126%x=20+1.281(4)=25.124\%\approx25.126\%

Thus the reported threshold is about 25.1%. Standardize first, use the standard-normal table, then translate back to return units.

Diesel prices in two cities

Jamnagar price XX has mean ₹90 and standard deviation ₹0.5; Surendranagar price YY has mean ₹92 and standard deviation ₹0.8.

  • Surendranagar at most ₹91: z=(91−92)/0.8=−1.25z=(91-92)/0.8=-1.25. The left-tail probability is 0.1056, about 10%.
  • Jamnagar above ₹91: z=(91−90)/0.5=2z=(91-90)/0.5=2. The standard-normal CDF is 0.9772; therefore the upper tail is 1−0.9772=0.02281-0.9772=0.0228, or 2.28%.
  • Jamnagar above ₹92: z=(92−90)/0.5=4z=(92-90)/0.5=4. The upper tail is very close to zero; the source describes this as roughly no stations.

Seed funding and normal approximation

This example is worked in the preceding approximation section: for 200 companies and a 20% transition rate, μ=40\mu=40 and σ=32≈5.66\sigma=\sqrt{32}\approx5.66. With continuity correction, at least 45 is approximated by Pr⁡(Y≥44.5)≈0.2148\Pr(Y\geq44.5)\approx0.2148, and at most 50 by Pr⁡(Y≤50.5)≈0.9686\Pr(Y\leq50.5)\approx0.9686. The count is binomial; the matched normal supplies approximate areas.

Key takeaways

  • Each usage mode, city, or time window can define a different random variable or parameter set.
  • Poisson counts and exponential gaps are two views of the same constant-rate arrival process.
  • Standardize normal quantities with their own mean and standard deviation before reading a table.
  • Continuity correction is needed when a discrete count is approximated by a continuous normal curve.

Test yourself on this module

Questions from these notes, one at a time, with feedback. Your weak topics are saved.

Set it up free Takes about a minute. Free plans work.