ࡱ> QP \p Sean Wallis Ba==xK`8X@"1Arial1Arial1Arial1Arial1dArial1dArial1Arial1dArial1dArial1Arial1Symbol1Symbol1Arial1Arial1Arial1Symbol1Arial1Arial1Arial1Arial1Arial1dArial1dArial1Arial""#,##0;\-""#,##0""#,##0;[Red]\-""#,##0""#,##0.00;\-""#,##0.00#""#,##0.00;[Red]\-""#,##0.005*0_-""* #,##0_-;\-""* #,##0_-;_-""* "-"_-;_-@_-,)'_-* #,##0_-;\-* #,##0_-;_-* "-"_-;_-@_-=,8_-""* #,##0.00_-;\-""* #,##0.00_-;_-""* "-"??_-;_-@_-4+/_-* #,##0.00_-;\-* #,##0.00_-;_-* "-"??_-;_-@_- 0.0000 #,##0.0000#,##0.00000000000000 0.0000%#,##0.0000000000.0000_ ;[Red]\-0.0000\ #,##0.000000000000000% #,##0.0000_ ;[Red]\-#,##0.0000\                 + ) , *            $  $    (         (  (      H  @  H  D        (@  @   L     (   ,    `analysis of trend,ZR&3  @@   HAexpected lower bound exactly x upper boundobserved sig levelsignificant (0.05)titlex linguistic eventsprobability margin e at least x F probability p binomial significances.dev. s mean m binomialdatay: F(x)  n: F(x-1)-F(x)    n: F(x-1)-E(y)    y: F(x-1)/F(x-2)    total c c(y)c(n) chi-square!probability of representativenesschi-square significancenp = F p(x) = F(x)/F(x-1)      Enter data in this row extend sequence as requiredchi-square contributions+compares observations of event at x-1 and x"#*CommentsAny recursively extensible linguistic phenomenon found in a corpus can be analysed provided that it exhibits at least two levels of addition.mzExamples include:F We need one additional event to find a probability p(x) = F(x)/F(x-1).34 56 :; <= ?@ AB p(x)+e(x) p(x)-e(x)Wilsonz(crit)zs scaled by F(x-1) Gaussian approximation (p + z/2n)/(1+z/n)   s = np(1-p)  z[p(1-p)/n + z/4n]/(1+z/n)   (for plot)g.o.f.z.s+z.s-$population proportion tests (simple) Wilson (1927) ReferencesYWe need two additional events to see if that probability changes between x = 1 and x = 2.9@ IJ ST WILSON, E.B. 1927. Probable inference, the law of succession, and statistical inference. Journal of the American Statistical Association 22: 209-212.Companion datasheet forG* NPs with x = {0, 1, 2, 3, 4, 5...} adjective phrases before the noun,  ,* VPs with x adverb phrases before the verb,  H* NPs containing x postmodifying clauses (i.e. modifying the same head), R* NP chains of x embedded postmodifying clauses (i.e. modifying subsequent heads), :* NPs with x co-ordinated adjectives before the noun, etc.  adjusted p' Ehttp://corplingstats.wordpress.com/2012/12/04/linguistic-interaction/>WALLIS, S.A. 2012 Capturing patterns of linguistic interactionB tM s !  :`8^ 0 P0 /0 T0&he0=0mT0Н03$l0s906]]7 zQaip0//, a1(;R;8 ;/T0t001w(v1zm00 E000߿0/t0\ /Percent0]Rء"Pjءء1</d 0&-`0>CRT00T0# /F-10 T0` /--tjw4vu`T0TvuK2L06kT0xuVgTvu  E9oS^_  dMbP?_*+%MJHP LaserJet 4050 Series PSlS 4dXXA4PRIV0''''HPP4(΍HSMTJ8HP LaserJet 4050 Series PSHPOrientRotate180FalsePostScriptCustomPageSizeFalseDuplexNoneInputSlotAutoHPDuplicateJobNameOverrideSWFWPageSizeLETTERPageRegionLeadingEdgeMediaTypeAutoCollateTrueOutputBinAutoStapleLocationNoneHPCoversOther_PagesTextAsBlackFalseJCLResolution600dpiHPConsumerCustomPaperTruePrintQualityGroupPQGroup_1JRHDInstalledJRHDOffHPDocPropResourceDataHPCabFileNameJRConstraintsJRCHDFullHPColorModeCOLOR_MODEHPPJLEncodingUTF8HPJobAccountingHPJOBACCT_JOBACNTHPBornOnDateHPBODHPXMLFileUsedhpmcpdps.xmlPSAlignmentFileHPZLSLHNHPSmartHubInet_SID_263_BID_514_HID_265HPColorAsGrayFalsePIUPHdA4 4 [none] [none]4P d?SEAN44dexcel.exeD:\program files\Microsoft Office\Office\excel.exe" dXX??U} } }  } m }  }  } $ } m } $ E;T0,h3 , T   pu,E6/Eu0GwT0 *  + ~ ? D@ D@ D@ D@& D '(A& LL( @(@(j@(3@(? % */8A/U@/@% m@%4@%?)   &'(((&&&    !.?  DD!.x쐺?  DD!.W]2M?  DD!.C˯`h?  DD!.? DD  "      33  44 4    +' >R ._? <DDDDDDD@@R .`aY? <DDDDDDD@@R .tgXز? <DDDDDDD@@R .x_p? <DDDDDDD@@R .ݜ܎\? <DDDDDDD@@ ,     s 0JѺ\? ]DDDD@DDDD@ADDDs  դ.g? ]DDDD@DDDD@ADDDs  U(? ]DDDD@DDDD@ADDDs -3? ]DDDD@DDDD@ADDDs s:{s? ]DDDD@DDDD@ADDD $.    + ! j?  D D ! آv?  D D ! Uc)? D D ! 5p_~?  D D !  9?  D D    ! (?! D D ! ?  D D ! M? D D ! Õ;ͬ?  D D !  Is0? D D     , ,,] ,$ GDD   s+-DD   s- ns"Bs-] ,, GDD   s+-DD   s- ns"Bs-] ,4 GDD   s+-DD   s- ns"Bns] ,<%GDD   s+-DD   s- ns"Bns 4 !!!!!!!  / !11(2>9^]?D D D(2IK=qg?D D D(2oI?D D D(2=aĭ?D D D(2A!?D D D  !21(2d٤?/Ŝ?   ! !9^]?4D$% M 3O&Q4$% M 3O&Q4FA1  3Of 3*?#M43*N#M! M43d" 44% S?M03O&Q D Probability p (Wilson intervals)'44eed9^]?N f?IK=qg?"y?oI?H sK?=aĭ?RV| ?A!?e ~v  <NMM? U ]``  MHP LaserJet 6L4C 4dXXA4DINU"4`f" dXX??3` .` . ` F,?.3d23 M NM4 3Q:  at least x FQi ;Q ;Q3_4E4D$% M 3O& Q4$% M 3O&Q4FAp 3Ofx 3*?#M43*N#M! M43d" 44% \U7M03OH&Q  Frequency F'44eee ~v # <NMM? p%<6]#`  "??3` 7` .` .-?3d23 M NM4 3Q:  probability pQ ;Q ;Q3_4E4 3QQ ;##QQ3_ O NM   d4J[T\?4 3QQ ;##QQ3_ O NM   d4J[T\?4D$% M 3O&Q4$% M 3O&Q4FA2 * 3Of 3*?#M43*N#M! M43d" 44% S}>M03O&Q H"Probability p (Gaussian intervals)'44eeT\?T\?dLg?dLg?WKY?WKY?:jj?:jj?~8s?~8s?e >@ Oh+'08@Th  Sean Wallis Sean WallisMicrosoft Excel@O՜.+,D՜.+,, PX|  ӰԺ1 analysis of trend  Worksheets 6> _PID_GUIDAN{0F9CEFE0-BC02-4BEB-A6ED-92D05E32C8AE}  !"#$%&'()*+,-./0123456789:;<=>?ABCDEFGIJKLMNORoot Entry FWorkbookSummaryInformation(@DocumentSummaryInformation8H