Continuous Random Variable- Introduction
1. Discrete vs. Continuous Random Variables
The primary distinction lies in the types of values the variables can
assume:
Discrete Random Variables: Take specific integer values.
o Examples: Number of successful sales calls (0–5), number of
defective radios in a batch (0–50), or customer arrivals in a
day.
Continuous Random Variables: Take fractional values within a
range (limited or infinite).
o Examples: Time between customer arrivals, the exact amount
of fluid in a 750ml bottle, or oven temperature for baking
pizza.
2. Probability Density Function (PDF)
While discrete variables use a Probability Mass Function (PMF), continuous
variables use a Probability Density Function, denoted as f(x).
Key Characteristics of the PDF:
Non-negativity: f(x) must always be greater than or equal to zero
(f(x) \geq 0).
Total Area: The total area under the curve of f(x) over all possible
values of X must equal 1.
Interval Probability: Unlike discrete variables where f(x) is the
probability of a specific point, for continuous variables, probability is
the area under the curve for a given interval [a, b].
o P(a \leq X \leq b) = \int_{a}^{b} f(x) \, dx.
Point Probability: The probability of a continuous random variable
taking an exact specific value is always zero because the area
under a single point is zero.
SIMILARITY B/W DISCRETE AND CONTINIOUS RANDOM
VARIABLE:
3. Cumulative Distribution Function (CDF)
The CDF, denoted as F(x), represents the probability that the random
variable X takes a value less than or equal to x.
Calculation: Computed by integrating the PDF from the minimum
possible value up to x.
o F(x) = P(X \leq x) = \int_{min}^{x} f(t) \, dt.
Tail Probability: The probability that X is greater than x is denoted
as \bar{f}(x) and is calculated as 1 - F(x).
4. Expected Value and Variance
The mathematical logic for these measures is similar to the discrete case,
but uses integration instead of summation.
Formula for Continuous Random
Measure
Variables
Expected Value
\int (x \cdot f(x)) \, dx over all values of x.
E(X)
\int (x - E(X))^2 \cdot f(x) \, dx over all
Variance Var(X)
values of x.
Note: Students are not typically required to perform integration from first
principles; instead, software (Excel) or statistical tables are used to find
these results.
Uniform Distribution
1. Definition and Core Concepts
A random variable is uniformly distributed if it can take any value between
a lower limit L and an upper limit U, and each value is equally likely to
happen.
Proportionality: The probability is directly proportional to the
length of the interval.
Probability Density Function (PDF): The height of the
distribution, denoted as f(x), is constant between the limits:
f(x) = \frac{1}{U - L} \text{ for } L \leq x \leq U
Outside Limits: f(x) = 0 for any value of x that is less than L or
greater than U.
2. Key Formulas
The following formulas are used to calculate probabilities and statistical
measures for a uniform distribution:
Cumulative and Tail Probabilities
Cumulative Distribution Function (CDF): Provides the
probability that X is less than or equal to x:
F(x) = \frac{x - L}{U - L}
Tail Probability: Provides the probability that X is greater than x:
\bar{F}(x) = \frac{U - x}{U - L}
Interval Probability: The probability that X falls between two
points a and b:
P(a \leq X \leq b) = \frac{b - a}{U - L}
Summary Statistics
Expectation (Mean): The average of the lower and upper limits.
E(X) = \frac{L + U}{2}
Variance: Measures the spread of the distribution.
Var(X) = \frac{(U - L)^2}{12}
3. Practical Example: Driving Time
The text uses the example of Nilesh Shah’s commute from Vadodara to
Ahmedabad, which takes between 100 minutes (L) and 140 minutes
(U).
Question Calculation Method Result
120
Expected Driving Time? \frac{100 + 140}{2}
minutes
Prob. of 130 mins or F(130) = \frac{130 - 100}{140 -
0.75 (75%)
less? 100}
Prob. of more than 105 \bar{F}(105) = \frac{140 - 105} 0.875
mins? {140 - 100} (87.5%)
Prob. of exactly 120
Probability at a single point 0 (0%)
mins?
4. Important Clarification: Exact Values
A common misconception is that f(x) represents the probability of a
specific point.
Zero Probability at a Point: For any continuous random variable,
the probability of it taking an exact, specific value (like exactly
120.00... minutes) is always zero.
Reasoning: Probability is defined as the area under the PDF curve.
Since a single point has no width, the area at that point is zero.
Intervals Only: You can only determine the probability for a range
or interval of values, not a single snapshot in time or space.