Basic STATISTICS wit Mi kenauaeas Requisites of a Measure of Cental a Furcon on vatnratce! Averages 48. Arhmet Mean 4, oo. Median 4.10. Parton Values 4.11. Mode 4. 2. Selection of core ee re vhoments 414, skewness 4.15. Measures 4-16, Kurtosis 4.17. ‘Statistica an Appropriate Avera A non 4.16, Siatstical Parameters fr Poisson Distribution 4.19. Statistical Parameters or Binomial Distribution te analysis 4.21. Types of Correlation 422. Coeficnt Parameters for Normal Disiibulon 4 24, Slandard Error of Estate 4.25. Regression Equation v 3. ‘4.1. introduction 42. Objects and Function® of Averages 4 Tendency 4.4. Types of Statistic yn 4.8. Harmonic Mi [42. intropuction “The most important objective of statistical of data so that it describes the overall level 1e whole set of data. era large group of raw data. One would like to know the critical value which represent llth mndency’ or ‘average value”. For instance, itis very stand the data concerning the income of millions of Indians. However, i! analysis is to determine a single value for the entire mass 1 of the group of observation and can be considered a representative value of th In other words, itis difficult for anyone to understand o remem| items in a series. Such a value is called “Central ter difficult to remember and under: itis said that in 2010 average income of the people in India was % 19,000 per annum, it will be easy for us fo guess the economic condition of most of the Indians. It is the average value which is called central tendency of the series, It is also called measures of location. Thus central tendency refers to a central value or a representative value of a statistical series. Definitions According to Croxton and Cowdon, “An average is a single value with in the range of the dala that is used to represent all of the values in the series. Since on average is somewhere within the range of data, it is sometimes called measure of central value.” 186ntire mass sidered a erstand or nts all the it is very owever, if e easy for 2d central a central ‘the data he range mn 187 wap Charks “An average is a igure that represents the whole group” 8M alloy and Smith “An average is sometimes called a measure of central tendency s values of the variable usually cluster around i” AND FUNCTIONS OF AVERAGES brief picture of entire data: The main purpose of average is to present a simple To genate deserision of the principal features of the raw data. As a result of it, data can ; understood. Comparison: Reducing the mass of data to one single figure facilitates comparisons To present Brn de cith f Meonean be made either ata point of time or over a period of time, For example, we ve ike percentage result of different colleges ina certain examination, say B.B A ig + fo-2010, and thereby conclude which college is the best or we can compare the pass percentage Eiaccame college for diferent time periods and thereby conclude as to whether the results sre improving of deteriorating iitaies statistical inference: An inference which is derived on the values calculated 3, To fort fom the sample is called statistical inference. Its helpful in knowing the parameters of population aking: Averages are helpful for making decisions in planning in various 4, ToAidin Decision-m: felis For example, a sales manager may need to know the average number of calls made per day by salesman in the field, A railway officer will required information regarding the ber of passengers carried by rails on the various passenger runs. Averages are average num ‘alible in siting standards, estimating and planning other managerial decision areas fo REQUISITES OF A MEASURE OF CENTRAL TENDENCY sy requirements to be satisfied by measure of central tendency Thefollowing are the fe 1. Simplicity: The fundam simple to follow 1 Representation: An average v should be based on all the observation of the series. 4 Rigidly defined: An average is left to the estimation of an observer fixed value it cannot be representative of a series. The bias of the investigator in such cases would considerably affect the value of the average. If the average is rigidly defined, this iaslabilty in its value would be no more, and it would always be a definite figure 4 Algebraic Treatment: The nature of the average should be such that it could be Saistical analysis ofthe data set. For example, it should be possible to determine the average Production in a particular year by the use of average production in each month of the year 1 8/-Sand Stable definition: A good average should have a clear and stable definition ental features of the average is that it should be easy to calculate and alue should be representative of the entire mass of data. It and it is not a definite and used forPROGNOAITY ag 6. Absolute Number: A good 8¥Cr08" should be A 1e observations (values) we m vrroa docs not sve a5 2 go0d avETaBS ‘observations should not yt rons: By extrem ‘of Extreme Observation: aaa : ae a ace Tus, «few very small OF VET IAB very large val e the value of a good SVC mpling: I two independent sample studies are uctuations of Sampling a : 4 La eof Fn Sol ot mately TE fo ak ge parla ele ee ce enquiries are made, cre is bound to es ile wn ene ee ein co cases this difference would be creat vite in es Comte values calculated but in abe eaten hn ich this difference, whiel a ae = ee than those in which its difference is more. rr eee: x ‘sampling’ is less, re consid Ais 0554 dis dev" [faa tees oF starisTiCAL AVERAGES isthe sified into two categories ower sitional averages ' 1 Mathematical averages pe pom Flow chart reveals their further categorisations Re ‘Averages i ositional Averages Half- Matas Aves poste ae prtbrate mean ee =a cen Partition values solut + + Harmonic mean Mode [fs. matHemaricaL averaces Various methods of calculating mathematical average of a data set are classified in accordance the nature of data available that is, ungrouped (unclassified or raw) or grouped (classified) data [46. aritumeric mean The most popular and widely used measure for representing the entire data by one value is aitheis mean. The arithmetic mean is the sum of the values in a group divided by the number of items in tht group. It is (a) simple arithmetic mean, and (b) weighted arithmetic mean 4.6.1. Calculation of Simple Arithmetic Mean - Ungrouped Data In this method, Arithmetic mean is calculated by adding the values of all observations and diving the total by the number observations. Thus, if x, x xy are the N values of a variate arithmetic mean () is given by 1. Direct Method: ¥ = “1~* ate oy S (Aee So ge on ee rithmetic EAN, Tela, | pst No of all the values of variable we of items or observations. ae Nisbet Method. When the mumber of items is very large, itis convenient to use the shor-ut tly aft snort short-cut method, we make use of an arbitrary origin, This arbitrary origin i tee good Inte med ean. We seck the difference between his A andthe actual value of ex ions the © made, | SoeH6t ied deviation and denoted as [Link], item. This ( S y sch other nt | see ano the ate omy tata |g aitmetc moan tions of f isant < ed mean. erin o tems from assumed, and i pumber of items oF obSCTVations. Nis | = — |g soLveD EXAMPLES ee! 5 example 4.1. The! following table contains the half-yearly bonus paid to 10 workers in a factory: \| e|ic2 [3] ai] Sesi aro a oa] | apreary | 150 | 200 | 300 | 650 | 280 | 190 | 400 | soo | sso | 220 Bonus (9 I | od ou the arithmetic mean. otto. Direct method: Lot the half-yearly bonus of workers by denoted by x Bae Half yearly Bonus x (in 3) : 150 2 200 | e 300 ae 4. 650 ata ‘ oo 6. 180 1. 400 arithmetic ; ns in thal : 10. N=10 dividing a te x, the ijn t- 2 \$ pratt method: Thus the mean half-yearly bonus is & 340. (Assume mean or A = 400)Z 600; N=10 ae A= 400; 54 aa goLvED | a = 490+ 2 =400-60 xa 400-10) Example 4.2- = 7340. sxatisies? cover, the direct method i “rhe mean is the same under botk te melds Moreover. 1 involves mor ae < a ah sues ofitems are large. This ca” De ‘quoided by employing short-cut method bs am ae ; , No. of stu 4.6.2. Calculation of Simple Arithmetic Mean - Discrete Series : solution. D 1, Direct Method. In a discrete series the values of the variable are multiplied by their respective. E n jucts so obiained are totaled. The total is ‘divided by the number of items, wich wencies and the er Prod fal ofthe frequencies. The resulting quotient 16 simple arithmetic average in a discrete, is equal to the tol of the series. Symbolically, K- where, | ‘fis frequency, fis sum of product of x and f, and ‘Nis the number of items or observations. | 2. Short-cut Method. When short-cut method is used arithmetic mean is computed by the followug, | formula \ L ee where X is arithmetic mean Hence, Ais assumed mean, dis deviation of items from assumed, and Nis ta uxnber of items or observations. aby thf ————————— ‘ation Method. This method is a variant 191 of shortcut method. It is adopted when the 2 as tiom the assumed mean have some common factor. The fllowing formulas used to calculated +3, Step Devii se en oc x Lfd’ roca where, { is arithmetic mean, Als ‘assumed mean, his step deviation, dis 24, ad h N is the m Note. Symbol d has been used to distinguish between umber of items or observations. the step deviation and the absolute deviation. Bh soLven EXAMPLES == Example 4.2. Calculate the mean of the following frequ statistics: jency distribution of marks in a test in ‘Marks: 10] 20] 30 | 40 so | 6 | 70 80 | yee] 3 | «| | | 9 (Boal heae | Solution. Direct method: Let the marks be denoted by X and the number of students by f- [Number of students | [OS Ss Sean 3 | 30 | 120 | 692. ee en eee Bare Tae tec Beene fd a pirect neo 3 x -30 -90 eth 10 : -20 - 120 short-cut 20 0 -10 a) 30 jiation 40 2 3 4 [Link] 4" 5 10 90 30 20 120 where 6 . i 6 . 30 CG ee. . a 40 80 Ais wrest eo . Ga js the mid N=Sf=50 Lfd=40, jis step deviatio BA ake 40+ 2 =408 tis ber Hence, average or mean marks in statistics = 40.8. Nis the number Step-deviatio method: SOLVED E Number of students Marks ay example 4.3. x f Sa sn 0, i 10 3 e 20 6 2 a a 10 1 -10 a 2 0 0 30 9 y 2 60 6 2 2 x z 5 Solution. 80 2 4 8 — _ (a) Direct m =D f=50 Lfd=4 ra fd’ X= Aves xh= 40+ a 10=40.8 N 50 Hence, average or mean marks in statistics = 40.8. 4.6.3. Calculation of Simple Arithmetic Mean - Continuous Series is the same as inthe The pracers fhe calculation of arithmetic average in a continuous seri of discrete series. in a continuous series the midpoints of the various class intervals are written down replace the class intervals. Once it is done, there is no difference between a continuous series aee, { series. In continuous series, the arithmetic = « cae tiven below ‘mean computed by applying any one of the three _ _ Ef 0 i pirect method X = =F" 20 00 > shorveut method X = a+ 2/4 3, Step devi tion method X = A+(2 xn) tere, { is arithmetic mean, 4s assumed mean, i mis the mid-point of various classes, his step deviation, os h Nis the number of items or observations [Bl SOLVED EXAMPLES Example 4.3. Find the arithmetic mean for the following data: | Marks ‘Number of students | 0-10 5 10-20 10 20-30 40 30-40 20 i 40-50 25 Solution. | @Direcimethod: X = = f " Marks Mid-values (m)| Number of students (f) fxm 0-10 5 5 25 10-20 15 10 1s 20-30 25 40 1000 30-40 35 20 700 the 40-50 45 25 1125 dst = se whet, combined mean of the two groups Mean of first series X, = Mean of second series N, = Number of observations in the first group | N, = Number of observations in the second group ote The above formula can be extended for calculation of combined mean of more groups as | ander: ‘Combined Mean Faas 1+ Np + N5+..+Ny fxd Case |: Calculation of arithmetic mean in case of cumulative frequency distribution =I0 [fl SOLVED EXAMPLES -10 Example 4.4. For the following data, calculate arithmetic mean, g Marks No. of students 20 L — 50 Less than 10 5 Less than 20 17 Less than 30 w Less than 40 4 Less than 50 49 Solution. A cumulative frequency distribution should first be converted into a simple frequency distribution as under: 0-10 5 | 10-20 7-5-2 20-30 317 = 14 30-40 41-31-10 | the ariel 40-50 49-41 =8[Marks , 5 25 0-10 180 10-20 Hy ra 350 20-30 a 10 350 30-40 40-50 45 E se rs=49 Xfm = 1265 E fm _1265 _ 55 $2 marks i-W oe Example 4: Above 0 Above 10 ‘Above 20 Above 30 Above 40 Above 50 Above 60 Above 70 Above 80 Above 90 65, Find the average marks of students from the following table: Above 100 0 Solution. The given data is in the form of more calculate mean we shall convert it into frequency distribution as given Marks 0-10 | 10-20 | 20-30 | 30-40 40-50 than type cumulati low No. of students 2-6: 65-55=10 55-43 = 12 50-60 43—28=15 frequency distribution Case Il: Caler Example 4.6 oat the average pc Pocket exp No. of stud Solution. C65 ution the data is obtained usis ow mean ae ofthe data i obtained using step-devation method asunder: [Mid-values (x)| Number of students (| d= *—8S ai pad’ 5 3 5 3 5 15 5 2 w 25 a 3 a 35 10 P Sy 45 12 -1 i 55 15 0 ¥ 6 2 1 " 5 c 2 12 5 2 3 6 95 8 4 32 Df=80 0 | Ba =-26 & fd’ (26) atl Wy xh |= 554-0 x10 = 55 —3.25 = 51.75 marks Case Il: Calculation of arithmetic mean in case of inclusive series. Example 4.6. The following table shows monthly pocket expenses of the students ofa class. Find cat the average pocket expenses. [ Pocket expenses (x): | 18-21 [36-45 | 4ess | se6s T [ [Link] students (): a | | 36 2a ania Solution. Calculation of arithmetic mean of inclusive series 1s the same as of exclusive series = Pocket Mid-values (x) Number of ae ped’ expenses (x) students ()) 18-21 | 19.5 | 8g <1 -8.80 22.25 23.5 33 -07 =23.10 26-35 305 55 0 0 36-45 405 36 1 36,00 46-55 50.5 20 2 40.00 «o) 60.5 2 3 36.00 L f= 164 Lf = 80.10Til )a3p + 8010) 19 = 305 +488 = 23538. (Ge anes sed arithmetic mean ‘Case Ill: Calculation of correct ae sc mcan ofa sercs of 40 items was callated by a studey ee Taam minread as@ 150. Find the Correct arithmetic neat ¢ wile ealeuating a ‘Mean, ‘Solution. As we know that, 2 Given that X = 265,N=40 ee zx 265= >. and SX (wrong) = 265 = 40 = 10600 Case V: Cale Correct value= 115 le 4. pe Tico _ boheme z) = 2X (wrong) + (Correct value) ~ Incorrec tne mean of the — ae 10600 zu a eet ‘he formula is: Case IV: Calculation of Missing Value Example 4.8. Find the missing frequency from the following data; itis being given that 199, is the average number of tablets for being cured: a cys 12-16] 16-20 ]20-24 [24-26 é [28-32 [32-36 ial Substituting EMME RIe ir > [aaa Shiels ed oes | Ms | 4 Solution. Let the number of person cured for tablet distribution of 20-24 is taken as ‘a’ ee | ect | it | 2 | Balink 10 | B | 4 the percentage uf | Md | -1 ENG Solution. | Z | ly | 0 ° be the mean mo 2 | a | 1 ; females employ, 2 18 Given that 3 SI 4 4 ase 20Set value) +a = 192(%0+4a)=152+a 4a-1920= 1728-152 2.08a= 208 208 ase V- Calculation of combined arithmetic mean Example 4.9. A cooperative bank has two branches employing 50 and 70 workers respectively. srerage salaries paid by two respective branches are 360 and & 390 per month. Calculate are imean ofthe salaries ofall the employees. - Solution. In this case, we have to calculate the combined mean of the two groups of emplowces The formula is: 1 = © 360; X, = 2390; N here Substituting these values, we get ¥,, = 300%50+390x70 _ 18000 + 27300 i 303470; 120 45300 Teo Ra The combined mean salary is 7 Example 4.10. The mean monthly salary paid to all employees in company is 1300, The male and female employee's monthly salaries are 1340 and 1260 respectively. Determine the percentage of males and females employed by the Solution. Let X, and X, be the monthly average salary paid to male and female employees, X12 te the mean monthly salary’paid to all the employees and let N and N, be the number of males and females employed by the company respectively Given that: X, = 1340; X, = 1260 and X,> = 13001340 N, +1260 N3 SOU SbeseeN | N: = 1300 (N, +N) = 1340N, + 1260 No 5 “oN, = 40%, M - = oh oof 1:1 or 50 per cent and 50 per cent respectively Hence, males and females are in the rati etic Mean ea SNe sen 3 the series is considered equally imporay jy ple average each item of H|ilems may not have equal importance, and some of them may ne rhe fundamental purpose of finding out an average jy comparatively more imporent ee coe ‘ae can, the central tendency of the many ve on ty reer calculated. This being so it is necessary that if some items ofa series ar my, ee than others this fact should not be overlooked altogether in the ealeltiog of an average rarer find out the average income of the employees of certain factory and if we simply addy Mreuey ofthe income of the manager, an accountant clerk, a labourer, and a watchman and dvd total by five, the average so obtained cannot be a fair representative of the income ofthese people Th soars that ina factory there may be one manager, two accountants, six clerks, and thousand lboury nd one dozen watchmen and if it isso, the relative importance of the figures of their income is nol he same, Similarly ifwe are finding out the change in the cos of living of a certain group of people ani tre merely find the arithmetic average ofthe prices ofthe commodities consumed by them, the average In the calculation of simy there may be cases where al would be unrepresentative Weighted arithmetic mean is an improvement over the simple arithmetic mean. Simple mean give ‘equal importance to all values while weighed mean gives relative importance to more significant items in the distribution. The term ‘weight’ is used in the sense of the relative importance of each item, The formula of weighted arithmetic mean is: rwx Iw Here, X,, indicates weighted average. I Sotven Examptes Example 4.11. Calculate weighted mean by weighing each price by the quantity consumed. [ Articles of food Quantity Consumed in kg | Price in & (per kg) Flour 11.50 i 58 Ghee 5.60 58.4 Sugar 0.28 8.2 Patato 0.16 25 Oil 0.35 m0 & aspee at ast taken ia egree ‘Quantity consumed mS (ver he) X ee Se 11.50 TS a ee 327.08 os oe 2296 ee oe 0.400 zo 035 en = DEAT) TWX = 403.436 2.55 — ighed mean price is € 22.55 Je 4.12. A bus runs ata speed of 60 kph over S0 kilometers; the next 30 kilometers at amp up the next 20 kilometers ata speed of 30 kph; and the Final run of 0 kilometers 1g a0 ph. What is the average speed i kilometers per hour? He Lat speed be denoted by X and time taken be denoted by HW, Now ealeulte the time nee Sot ce of diferent speeds. ee Lee so tme 1d of 60 kph = 50 the average Been cover 50 kms ata spocd of 60 kph = 50 minutes Toetsken to cover 30 kms ata speed of 40 kph = 45 minutes mean gives Frociaken to cover 20 kms a speed of 30 kph = 40 minutes a Faeiaken to cover SO kms ata speed of 25 kph = 120 minutes me “mange these data in tabular form, then ‘Speed in kph () Time in minutes (W) WX sp eal 60 | 50 3000 40 45 1800 30 | 40 1200 25 | 120 3000 onsumed. 7 | DW=255 YD WX= 9000 _| — — SWXx _ 9000 ZWX _ 9000 _ 55 29 X= Sw tas Wighted mean speed is 35.29 kphEEE | 20: 4.6.6. Merits of Arithmetic Mean y used measure of central tendency, There aye May is the most popularl ‘The arithmetic average 1s seat, the following merits because reasons for its popularity alculate. 1. Itis simple to understand and easy 10 © Therefore. itis the mst represen i m es. 2 In is based on all the observations of the ser d valuc fre is a certain value; it has no scope for estimated v a 3 ett erie algebraic treatment and we can use it for future mathematical CAedatig 4. It is capable of r alge i Briere by fluctuations of sampling and ensures stability : calculations 5. It 1 otic mean ean be tested for ils accuracy’ as a representative value ofthe srg etic ; fo stable and certain, arithmetic mean can be easily used for comparison, Arithmetic Mean ees which cannot possible exist, eg it ed results rithmeti sometimes gives most absur ; aan find out the number of children per family, and if we use the arithmetic ay ‘ itis quite likely that we get the average as 3.4 children. Obviously the result is absu itis ly that w 2 divided in fractions. 2 ae are mean cannot be calculated accurately for unequal and open-end lass intervals either at the beginning or end of the given frequency distribution, 3, Since the arithmetic average is calculated from all the items of a series sometimes thea items may considerably affect this average, particularly when the number of items is ny large. For example, if the income of a shopkeeper is & 1,00,000 per month and the incom of his three assistants are % 2500, % 3500 and % 4000 per month respectively, the averag income of this group would be 27,500 per month. This is not at all a representative figure 4. The fact that the arithmetic average cannot be calculated without all the items of a seriesca also be said to a demerit. If out of 1000 items the values of 999 items are known the arithmeis average cannot be calculated. 5. The arithmetic average gives greater importance to bigger items of a series and lesser importane to smaller items. It has an upward bias. 6. Arithmetic mean cannot be used to qualitative data such as honesty, beauty, loyalty ete [7. Geometric mean In many business and economics problems, we deal with quantities (variables) that change ova period of time. In such cases, the aim is to know an average percentage change rather than simple average value to represent the average growth or declining rate in the variable value over a period of time. Thus we need to calculate another measure of central tendency called geometric mean (GM) In the previous sections, we have seen that average can be calculated differently for various tyes of distributions. Likewise, geometric mean can also be calculated for three different types of sets These are: \ Irdiviiu eries or ungrouped data; 2, Discrete frequency dis 3. Continues frequency distribution ribution; and where salues of ‘oki | Exe Sol ro}cs, eet Nerage surd. 4 ended normal is not comes verage igure es can metic era nple dof ) ics. ti 474 ie mean i defined as the nth rt ofthe product of items ofa ay gad 818 10 be calculated it would be equal othe cube ros st iy be womete meas of 6, 9, 12 and 16 would be the 4%" 60 ee = cries. Thus if the geometric the product of these figures, Foot of the product of these four GM. = fn) Xx, xy [Link], = x xy x) jee GM. stand for the geometric mean, nfo the numberof items and «” ihe variable 19 ng logarithms on Both the sides ofthe above equation, we have and £9. ot forthe 1 log GM. = | (log; + log x, + log x, +1. logs) log GM. ! a4 a GM= Atos ( 1 Sos } (aa Jf SOLVED EXAMPLES example 4.13. Calculate the geometric mean from the following items: 133, 141, 125, 173, 182 Solution. Calculation of geometric mean | Size of the item (x) log x 133 2.1239 141 2.1492 125 2.0969 173 2.2380 182 2.2601 5 Y log x= 10.8681 | Wehave, ¥ log x= 10.8681 and n= 5 By applying geometric mean formula we get, (12 GM, = Antilog | — Ylog x; (tags = Antilog ) = Antilog (2.1736) = 149.14204. Example 4.14. The respectively: What is the average rate of growth per ai Solution. Ifthe growth in the beginning is annual rate of h for a factory for 3 years if 3%, 8% eo nnum for this period? and ay, 100, then for 3 years, the growth i 2 given by Proms the ‘Rate of Growth at d ee fret end of year (x) 1 105 2 108, Fs io 112 (La ce By applying geometric mean formula we get, The required average growth rate is 108.2 - 100 = 8.2. Therefore, the average rate of guy percentage per annum is 8.2% GM. = Antilog E Dlg s] = Antilog: [s 138 = Antilog (2.0346) = 108.2. [ee We have, 5f By applying § 4.7.2. Calculation of Geometric Mean — Discrete and Continuous Series In the case of a frequency distribution, geometric mean of 7 values x,, x, x, occurring with frequencies /;, fy... /, respectively is given by x, of a variae GM.= (4/4 xf)” where, N=/,+f,+f,4, +f Taking logarithms on both the sides in the above equations, we have eile’ f cee log GM. = 5 (flog x, +f log x, +f, log x, +f, log x, Marks: No. of tua pe log GM. = 5, DCJ, log x) Solution. | fi data ae ) GM. = Antilog | = ))(/, tog x) | ‘a ) Similarly, the geometric mean of the interval, in the usual way. for a continuous series can be obtained by fin ‘out the mid-value