Measures of dispersion, Applied Statistics

 

Other Measures of Dispersion

In this section, we look at relatively less used measures of dispersion like fractiles, deciles, percentiles, quartiles, interquartile range and finally quartile deviation. Quartile deviation is also referred to as semi-interquartile range.

Fractiles are similar to percentages. If we say that in CFA level I, the pass percentage was 35, we understand that out of every 100 candidates who appeared for the examination, 35 candidates passed the examination. Similar to this, in a frequency distribution also there are always some elements which lie at or below the given fractile. For instance, the median is 0.50 fractile as half of the data is less than or equal to this value. Further in any distribution, 25 percent of the data lies at or below the 0.25 fractile. Interfractile range is the difference between the values of two fractiles. Therefore, this is also necessarily a measure of dispersion between the two fractiles in a frequency distribution. Depending on the number of equal parts into which we divide the data we call them as deciles, percentiles and quartiles. A decile is a fractile which divides the data into ten equal parts, percentile is a fractile which divides the data into 100 equal parts and finally a quartile is the one which divides the data into four equal parts. While computing fractiles we ought to arrange the elements in an increasing order.

Interquartile Range: Since range, interfractile and interquartile range typically from what we call as distance measures of dispersion, the interquartile range looks at how far one should go from the median on either side of it before one-half of the data points are included. As we have seen above, quartiles are fractiles which divide the data into four equal parts such that each quartile consists 25 percent of the data points. The quartiles are the highest data points in each of these four parts. Interquartile range is the difference between the last value in the third quartile (usually denoted by Q3) and the last value in the first quartile denoted by Q1.

Quartile Deviation: It is the difference between the third and the first quartile divided by two. It is expressed as Q.D.

Q.D.

= 1077_quartile deviation.png

where,

         Q1 = first quartile

         Q3 = third quartile

Posted Date: 9/14/2012 3:01:58 AM | Location : United States







Related Discussions:- Measures of dispersion, Assignment Help, Ask Question on Measures of dispersion, Get Answer, Expert's Help, Measures of dispersion Discussions

Write discussion on Measures of dispersion
Your posts are moderated
Related Questions
Differentiate between prediction, projection and forecasting.

In a three-cornered paint ball duel, A, B, and C successively take shots at each other until only one of them remains paint free. The three paint ballers have different probabiliti

This question explores the effect of estimation error on apparent arbitrage opportunities in a controlled simulation setting. We simulate returns for N = 10 assets over T = 30 year

The investor has constant wealth 1 and is o?ered to invest in shares of a project that either gains 3=2 or loses 1 with equal probabilities. Therefore, if the investor obtains sha

If the data set contains an odd number of items, the middle item of the array is the median. If there is an even number of items, the median is the average of the two items. If the

Assumptions in Regression To understand the properties underlying the regression line, let us go back to the example of model exam and main exam. Now we can find an estimate o

Poisson Distribution The poisson Distribution  was discovered  by French mathematician simon  denis  poisson. It is a discrete probability distribution. Meaning : In bi

Let X, Y, and Z refer to the three random variables. It is known that Var(X) = 4, Var(Y) = 9, and Var(Z) = 16. It is further known that E(X) = 1, E(Y) = 2, and E(Z) = 4. Furthermor


In PCA the eigknvalues must ultimately account for all of the variance. There is no probability,'no hypothesis, no test because strictly speaking PCA is not a statistical procedure