Scribd
Upload
3 views3 pages

Part 4

This document reviews the concepts of random variables, focusing on discrete and continuous variables with examples. It explains the construction of frequency tables, bar diagrams, and histograms, emphasizing the importance of using relative frequencies and densities. The document also highlights common mistakes in histogram representation and the significance of area over height in accurately conveying data distribution.

Uploaded by

Tonmoy das
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views3 pages

Part 4

This document reviews the concepts of random variables, focusing on discrete and continuous variables with examples. It explains the construction of frequency tables, bar diagrams, and histograms, emphasizing the importance of using relative frequencies and densities. The document also highlights common mistakes in histogram representation and the significance of area over height in accurately conveying data distribution.

Uploaded by

Tonmoy das
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Random variable

To understand this chapter, we have to review bar diagram and histogram from
STAT H-101.
Example (discrete variable):
Let 𝑋𝑋 = number of accidents in an area in a week. Let 𝑛𝑛 = 50.
Data: 2, 4, 1, 0, ⋯ , 1, 3.
Frequency table:
𝑥𝑥 Frequency Relative
Frequency
0 6 0.12
1 14 0.28
2 12 0.24
3 10 0.20
4 6 0.12
5 2 0.04
Total 50 1.00

Bar Diagram

• We can call it bar diagram though we used lines instead of bars. We should
NOT call it a line diagram. Line diagrams are different.
• We used relative frequencies instead of frequencies (to compare with
probabilities later).
• Height of a bar (line) is proportional to the relative frequency.
13
Example (continuous variable):
Let 𝑋𝑋 = lifetime (years) of an electric component. Let 𝑛𝑛 = 100.
Data: 2.85, 0.54, 1.29, 11.23, ⋯ , 3.55.
Frequency table (class intervals with unequal widths):

Class Frequency Relative Density


Frequency = rel. fr. / class width
0–1 39 0.39 0.39
1–2 24 0.24 0.24
2–4 24 0.24 0.12
4–6 8 0.08 0.04
6–8 3 0.03 0.015
8 – 10 1 0.01 0.005
10 – 12 1 0.01 0.005
Total 100 1.00 Total is meaningless

Incorrect Histogram

In the above histogram, “height” of a bar is proportional to relative frequency. The


class ‘1 – 2’ has relative frequency 0.24. The class ‘2 – 4’ also has relative frequency
0.24. These two classes received bars of equal heights (though the class ‘2 – 4’ has
more width). Why is it wrong? To understand this, a vertical dotted line is drawn at
𝑋𝑋 = 3. Even if this dotted line is not drawn, people will imagine it subconsciously.
14
Then, people will think that there are equal number of values in ‘1 – 2’, ‘2 – 3’ and
‘3 – 4’. That is, they will thank that 24% of all values are in ‘2 – 3’ and an additional
24% are in ‘3 – 4’, which is clearly not the case.
In the following histogram, “area” of a bar is proportional to relative frequency. This
means, height does NOT represent relative frequency and we should NOT label the
𝑦𝑦-axis “relative frequency” or “frequency”. Height of a bar represents ‘density’
which is defined as:
relative frequency
density =
class width
That is,
relative frequency = density × class width
(area of bar) (height of bar) (width of bar)

Correct Histogram

Here, people will think that 12% of all values are in ‘2 – 3’ and an additional 12%
in ‘3 – 4’. This is a simplified extension of the fact that 24% of values are in ‘2 – 4’.
We discussed an example with unequal class intervals. When all the class intervals
are equal, density is proportional to relative frequency. However, we should use
‘density’ in both cases to maintain similarity and to make probability discussion
easier.

15

You might also like