A competitive book of egricshon, To Hal Sausues was a} John G The purpose of randomization eld expen S10 Contra “variance Unit ess measure of dispersion is Coeficient of Vario is computed by S./Meun)« 100 Crieal Differcce (CD) i clculated by 2 MSE) x1 Expression for deg fom in RED i (1) (1) ‘The range of probaly les btvesn: 0107 Minima dears ofrsdom far eno in ANOVA is. 12 “T tests appliable when nur of weatnent is 2 's used fr dstbution of Chi square test The first sip of sunning the data i Clasifeation Prob of possible event is alas. Zero To find the average height of plants We Shou WS: Aritimetc mean. When sample siz is sll and D se 7 tet For sting the significance of correlation ceticient use 7° tet Lan square design suitable or comparing 012 retments Ine el experiments the commonly sed design i: RAD The eression coeficicm es between. = 10+» The colton ofc! es betwen: -1 0 +1 “The student est was discovered by: WES. Gasset The vale ofS (chi square) alas lis between: Oto Most commonly used mcasre of cena tendency 8. Arithmetic Kass snnnss ¥ Measure of dispersion which is considered as best Standard deviation 4 Which is not a variation. Theale of standard deviation may vary betwee 0 » Correlation is used 19 analysis the behavior of Tw or more ariable, The RAD is available fora wide range of ucaents 240 24 IFobserved frequency is equa 1 expected one than The value of To geting the higher precision i required: Larger samples YP est cam be used for testing the significance of Severaldfferences mum ple size for using tet shouldbe: 50 389 absolnte measure of dispersion Coefficient of ty. Missing plot techrique 1 Two study effet of 90 oF more sizes Split Plot Design In bts FF QU como: the dispersion: Range 8 ier mee of Y-ifall he var values ae neue he a Positive nder the SSSUMpLION for ahs of population should be Norma 4 The anahsis of yanance technique i wel for means of more than 0 popltion, V_The design whe ounber of repicin i lo andr of treatments Latin square design (8D) ‘ 1 For tesing the independence of 0 bse os ut test 4 Concation of continaty in 22 comings ble tule as ‘when expt equncy onc Leshan § When two variable moves in the sae don conan ssid tobe: Positive When two vanables mores in he pst distin colton sad to be Negative In the case of pene maine Conlon he cot of correlation willbe -1 Probability ofan event is. umber hing nce, 0 ba ot Sugden pa nda dove i be Yamane. the pent omparing the practically zero In describing the amount of varaion ina pomtion & mesie often used is Coefficient of aration ifthe fertility variation in the eld in two direcins a it Angles. which ofthe eyprimetl dans sue SD ¥--Anhmtic mean is most commons we because His bsed on at observations bo ¥ To testing the omogeneny of conan eect ws square tet th ropes itera a ¥ To reduce the experimental era! used: More replication & ws of let 390standard deviation is expressed % of mean is known as Ey value of CV is greater it means mote variability or lege Moment. + Forge » distance + The moment in statistics are used 10 describe the anon, hurts of fequency disinbution like Cental tendeney (n ) Yartance (2, skewness 3) and kurtosis (4), © Ina symmetrical distribution all odd moments ie 41.93, woulg always be zero because positive deviation & nese deviation af ‘summetry ane exactly balance oF egal Cena moment origin * Usually cakulation of central moment takes the more ine de to reason that mean muy no be in whole number therefore calculation ofthe central momen ment and wath lp nt UNSE.X-AY' Central moment: + The F moment stout mean (x) is defined as. = UN DFX-X)" 1 Value of w= 393 + ws INTROS +} cemal man Besa dispersion, Neste so lc of 3 cena mont + y= INTE, Chek u never + hthssands orn 5 Value of 4 central moment ad + noe UNE, OX) Sia "There are two typeof fequeny dsttuinn, 1 Symmetric frequency distibunes ‘+ Mean= Mean Made, Asymmetrical fequens dst Mean Meme Mode Quarts reno equ isan fom mtn Q\-Md2 Md-Q), ‘Curve is more icin in one si thane. The sum of positive deviation fo iin ntl 0 ‘he sum of nega deviation Hom te metan, cree at ee cegeoee 2 * Positive skewness = Mean > Median Mode, a ae Slam | Measures of skewness: 3944 ohol meri ji ~ Taken Te of Sic agg eek pry 1 SS Mean Mode p= probability of success | 1B probability oral or) t pra= 18D. rprcen ty 1 % Bue on moments 1 fiparethe ramon ert + 'Shevsness (fh) = 17 6. xu properties = yorret rerietc EAN (KH) OFBD= yp + kaos = + Variance ofS = Be soynh3 J stowness 9 a For symmetric distribution: cis 30 oe a + Standard Devianon = RR *8020q ok * + incase of BD / en 3 \ Mean> variance 07) poisson distribution PeSprscov ered by SD, Poison + Refers to degre of fatness or peakness ofthe frequency 3 eit: ‘= Tis denoted as itis. iting case of B.D such tha ‘Three types: + Ntendsto © = Leptokurtic narrow peakbase): Ep isd 20 = Mesokurtc(normalcune):——=3 + and np =m + Phaykuric (Mat peaRbrond base) <3 Disebation: Theoretical dsinbutions ar two types a Digerete disibuvon +) Xeomtake onl integer vas ike 0.1.2.8 +E Binomial distribution, Poisson distribution Continuous distibation + Xean ake all posible values ina given range + Eg Norm 1 Binomial distention: + Gen by James Bernoulli in 1700 A.D + Bernou ia ra s known as Bernoulli a i it ha nls 10 ‘outcomes namely succes and failure with probability pa @ “hefore a Variable x” sud follow aFosson dion i sumes ies and ts proba nas non ens cals negative v x Mere is called the parameter of PD ¥ disalways greater than 70 ¥ X= number af susess ¥ e= 2718 respectively + pranl Examples of PD: . Definition ‘Number of deaths from a diss ‘+ Arandom arable “x sido have binomial disibution i + Number of defective mates, enh : wrong calls eens 4 assumes on non-negative Values and its probability mass Function a, Nas of wrong il Is given by ropeties of P.D: = Po= tain PS g?* Mean (u): is © X=012— + Variance (1 + f= numberof wat os 395 "s A competitive book of agri Since Nene Vatance OR EON ee, PLD is sei in theory of games wating tine, problems of business 4 PD tends to normal distbation when i. tends to Coefficient of P.D: = nen * Be 34 3. Normal distribu + Discovered by de- Moivre in 1733 also called Gaussian ‘dstrbution Definition: * Arandom variable x” ssid to have a normal distnibutio ‘parameters mean and variance and its pro is defined as: Pe 1 with ity density function el) * oV nm Properties of normal distribution: "The curve of normal distribution is ell shaped ‘tis symmetric about the mean ‘Mean = Mode = Median (are ual). Quarles are equi distance from median: 0-0: = 0.-0, Coefficient of skewness for normal distribution is: Zero Coefficient of kurtosis for normal distribution is: 3 Range of normal disiibution is from: = tobe But practically range equal io 6 deviation about mean: Se le deviation: 236 Importance of normal distribution: thas gota large application in statistics quality control and important area of statistics used in industries for setting control limit It studies the relation or association between two variables. ‘Two independent variables are uncorretated ‘The measurement of coreation i called the correlation eo- efficient () or corelaton index, which summarizes in one Figure the direction & degree ofcoreaion. ‘Range of correlation are varies between +1 10-1. 397 4 both he varabls ray vail easing ee ieee own asthe negate coreg 4. Simple partial and multiple corms ‘Simple comcation: When ons were {Partial coneiaon: More than tray consider ony tobe nrc + Multiple coraton thre erm i asd simuancous Linear and nen linea cordon: + Ifthe amourt of change non vrai ex exact ratio the amount of change inte eae omens near correlation + te amour of change in ail dea he cons io the amount of change nor ancl shown sno dna correlation Inthe mos ofthe pail suis wed so ier relationship between variables. + Absence of any relatonshi correlation coefficient will : ables ait ables ae ued aril vale of ins Term regression firs sed by Sir Francis Galton + The repression ams i a mae memes ay Original ine of regression + Range of regressions varies between: #2 18- en isi aries Wen te coca toms oa Steam en ites (CO SEN ian gatas Variance (6) Types of transformation: = Spare rt anor yy tenet Used when countable dit data follow Poisson —_2 i d book of agricatng Serie adore a — + Eg: Transformation =» Logarithmic transformation oe Vee when vaio i the counable nas arg and values ae aso lee + Eg Trasfomation oe x orloe +1 © Angular or arsine transformation Jed ford following boa dsibuton and when dat arn percentage based on count ales The perceage are transformed he doce + Eg Sin= pio spate, Ay afement about he population ante Test of hypothesis: Nal notes 1A ement bout parsers whi be rected afer testing now al hypothe + Given by RA. Fisher + tisdenved by + uisthehpotiess om iference 1b Attentive hypothesis ‘Tiss the ypothesscomay tothe nl hypothesis Denote by Hh yes of era a Type Fervor se dete by Alfa a + Rexcting H, wens o > Type Menor +" Denoted Beta) + Accepting Ho when tis as. rejctng H, when i swe + tisiore severe than pe enor Level of significance ing Hy when iis fase ‘The maximum probability of commiting type I error is known as level of significance denoted by Alfa + Generally we tke 5% (eld ex.) 0 Degree of freedom: ‘is diferenc in vot munber of tems (n)& linear constants. evel of significance + DAR = N- Tota numberof constrains DE = NK (otal wo of constrains) ‘Tes of significa 399) CoP st used by Kart Fearon 11s used for enumerated oF on yr Mendelian or population yneics sabe dan Range: 010°" Fe Ores sons for applsing x test “otal sumplesze®80 Epsted frequency of consti and xe, class shouldnt thai n w «|The sample obseration shuld eindgenen agotetion of Chi-square Hr eo est the goodness oft {For esting the hypothe about the atin for various class {Testing the independence of mo ates {For detection of linkage «Tasty gene fequencisin popsaion pasts 1 Test of significance of diferenc eveen tvomean by Zites student test + Given by RA. Fisher, Used when sample sei large> Student t-test Given by WS Gost 1908 {Used when the sample sie imal (30 ‘Base principle of fed esperimentati Replication: : +P aepetiton of eatment under ines non 5 repeion aera remem experi ere tp of iene ne field + Itineeass precision Randomization: + AMlcation of weatment othe ae random proces + Tisminns eran © Local eonteoleror conto: to estimate experimental emo ernest is = tral at iret Local one age cia a eet Te ac ne re ot nate a ess " 23 a crm ete mee seni nr poet ‘Treatment: 400oon Made cester broncane res weksounas + Bop Varieties, manures. culivation practices. methods of ye ‘treatment. insecticides (ar treatments used in cae of fel experiment Experimental unit The group of materia to which treatment is applied ina single teal of weaument + Eg. Aplotof land a patient in hospital te Experimental err The variation inthe data due to uncontrolled factor isknown as perimental eror ANOVA ‘+ Analsis of variance The concept by RA. Fisher + Ins stastcal technique of partitioning the oa variation ino ‘component varations & computing them by F test, ‘Experimenta design, "A local structure of an experiment hat elps in obtaining results ith some precision, 4 ERD (completely randomized design: + Wis regarded as one way clasiiation and mo way control or clamnation + Applied when the experimental material is limited and homogenous. sch as the soil in Pot experiments + This provides the maximum number of degree of freedom for ero. fora gen number of treatments as compared o other design Local contol mot used in CRD, Error degre of freedom in CRD: New? b bck design | {Teor claniicaton ad one may coed + Thsdosign appropriate when the fer ‘one direction ons 1 used upto 20 treatments without appreciable lass of BD hot nr ks mst sequal o ras he nerf ‘ephicatwons fived for ceach treatment “4 The numer of pts i eah block ‘em, should be equal othe uber of 401 Spor nds more socuate Teale fromozenous blocks, itis mos commonly it dsp in hus dessn ilies al beter bene ee caperimentation Ys ppl 4 ror dew of reedom in RED (exe {sp (Latin square design) SP ce wa lastiaton to way conte enr 1c erm square hasbeen sed in SD hos dud columns are equa This design used when tity gdm nto dies + Thi eet ein medi Because the munimum eno ders feo sel {shouldnt be use om sete mabe of oma ‘amber of replications = Naber of treatm ‘Number of ows = Number o alum » Number f treatment, ‘Randomization of treatments done in sucha wa that each treatments oscurs once and onl once in each row an each column Error degree of feedom in LSD: (we) (2) 4 SPD (split plot desig) ‘+A layout in which one set of teatments is asgnd oar plots ‘alld main plots ino which a replat sit and another st ‘of treatments to sub division of the main ls ald sab pas is termed aa split plot design. “This design used under following condition 1s) When factors ofthe ifferen tue are stein sme experiment tat is lve fone fico equ e 38 ‘compared to other factor eg, Depth of ploughing am Aoae® Tevels, date of sowing and varieties, vars an mops tmigition level and vant. (i) When all factors aren't importa. . (ay When he els of ane cn ta lager ise ‘compared with the levels of " . oy Whe ant to ty on fan wi er PS ‘compared to other factor athe Hm sib pas €sePb(strp plat ee: lane ree sere ev wx hee ist ks cat i a ‘andomiing nh 402,