Binomial distribution, n = 20, p = 0.8
For X ~ Binomial(n = 20, p = 0.8) the mean is 16.000, the variance 3.200 and the standard deviation 1.789. The most likely number of successes is 16, with probability 0.2182.
How to work it out
- 1
A binomial counts the successes in n = 20 independent trials, each with probability p = 0.8 of success.
- 2
The probability of exactly k successes is the number of ways to choose them times the probability of one such sequence.
- 3
For example, the most common value k = 16 (the nearest whole number to the mean):
- 4
The mean and the standard deviation come straight from n and p.
- 5
np or n(1āp) is below 5, so the normal approximation would be poor here; use the exact binomial probabilities.
LaTeX source of the formulas on this page
\[ P(X=k) = \binom{20}{k}\,0.8^{k}\,0.2^{20-k} \]
\[ X \sim \mathrm{Bin}(20,\ 0.8) \]
\[ P(X=k) = \binom{n}{k} p^{k} (1-p)^{n-k} \]
\[ P(X=16) = \binom{20}{16}\,0.8^{16}\,0.2^{4} = 0.2182 \]
\[ \mu = np = 16.000,\qquad \sigma = \sqrt{np(1-p)} = 1.789 \]Binomial(20, 0.8) probabilities
| k | P(X = k) | P(X ⤠k) | P(X ℠k) |
|---|---|---|---|
| 9 | 0.0005 | 0.0006 | 0.9999 |
| 10 | 0.0020 | 0.0026 | 0.9994 |
| 11 | 0.0074 | 0.0100 | 0.9974 |
| 12 | 0.0222 | 0.0321 | 0.9900 |
| 13 | 0.0545 | 0.0867 | 0.9679 |
| 14 | 0.1091 | 0.1958 | 0.9133 |
| 15 | 0.1746 | 0.3704 | 0.8042 |
| 16 | 0.2182 | 0.5886 | 0.6296 |
| 17 | 0.2054 | 0.7939 | 0.4114 |
| 18 | 0.1369 | 0.9308 | 0.2061 |
| 19 | 0.0576 | 0.9885 | 0.0692 |
| 20 | 0.0115 | 1.0000 | 0.0115 |
Summary
| Quantity | Value |
|---|---|
| Mean np | 16.000 |
| Variance np(1āp) | 3.200 |
| Standard deviation | 1.789 |
| Most likely value | 16 |
| P(at least one success) | 1.0000 |
| P(no success) | 0.0000 |
| Normal approximation usable | no |
Questions
What is the mean of a binomial with n = 20 and p = 0.8?
μ = np = 16.000.
What is the standard deviation of Binomial(20, 0.8)?
Ļ = ā(np(1āp)) = 1.789.
What is the probability of exactly 16 successes in 20 trials with p = 0.8?
0.2182.
What is the probability of at least one success?
1 ā (1 ā 0.8)^20 = 1.0000.
Can I use the normal approximation for n = 20, p = 0.8?
Not reliably: np = 16.000 and n(1āp) = 4.000, and one is below 5.