Alternate Dispersion Measures in Replicated Factorial Experiments

Size: px
Start display at page:

Download "Alternate Dispersion Measures in Replicated Factorial Experiments"

Transcription

1 Alternate Diperion Meaure in Replicated Factorial Experiment Neal A. Mackertich The Raytheon Company, Sudbury MA Jame C. Benneyan Northeatern Univerity, Boton MA Peter D. Krau The Raytheon Company, Tewkbury MA Abtract Any of everal tatitic traditionally are ued to detect diperion effect in the analyi of replicated factorial experiment, including the within-run tandard deviation, the natural logarithm, variou ignal-to-noie ratio, and other. Thi tudy examine the relative performance of each approach uing recent experimental deign, with typically producing the bet reult. An alternate approach, baed on the abolute deviation from the within-run mean, alo i hown to increae detection but at the expene of uncontrolled fale alarm. Introduction While deigned experiment hitorically have focued primarily on optimizing the mean of one or more repone, they increaingly are being ued to identify factor that affect repone diperion. Example include increaed interet in variance reduction a a primary objective, increaed emphai on proce and product robutne, and work on deign for etimating variance function (e.g., Box (1988), Box and Jone (1992), Byrne and Taguchi (1986), Davidian and Carroll (1987), Phadke (1989), and Vining and Schaub (1996)). In order to detect difference in repone variance due to factor effect, any of everal tatitic traditionally are ued in the analyi of replicated factorial experiment. Thee include the within-run tandard deviation, the natural logarithm, the within-run variance 2, variou ignal-to-noie ratio, and other. Thi heightened emphai on repone variance increae the importance of undertanding the relative performance of alternate poible meaure for detecting true variance effect in typical experimental cenario. The current tudy alo i motivated by the uggetion that the bet meaure for detecting difference in ome value of interet might not necearily be one of the mot familiar tatitic for etimating that parameter. We examined the relative performance of conventional and alternate type of diperion meaure in recent experimental cenario, a well a another type of diperion meaure motivated during the coure of thi tudy by a deire to increae power by preerving aociated degree of freedom. 1

2 Alternate Diperion Meaure in DOE Reult ummarized below indicate that tend to be the bet traditional meaure for detecting variance effect, while larger-the-better and maller-the-better ignal-to-noie ratio tend to be quite poor. Two alternate tatitic uggeted during the coure of thi tudy, the abolute deviation from the within-run mean, y i,j - y i, and an approximately normalized tranform of thi tatitic, y i,j - y i.42 increae power in everal example over,, and popular ignal-to-noie ratio, although at the expene of uncontrolled type I error rate. Traditional Diperion Meaure in DOE The mot common meaure ued in replicated deign to identify factor and interaction affecting repone diperion are well-known in the literature (for example, ee Box (1988), Box, Hunter, and Hunter (1978), Montgomery (1984), Myer and Montgomery (1995), Schmidt and Launby (1995), Taguchi (1986)) and include: the within-run ample tandard deviation,, of all repone replicate at each et of experimental condition, the within-run ample variance, 2, of all repone replicate at each et of experimental condition, the logarithm or natural logarithm of or (+1), and Taguchi' three mot common "ignal-to-noie" ratio (i.e., "nominal-the-bet", "maller-the-better", and "larger-the better"). The within-run tandard deviation and variance 2 are direct etimate of their theoretical counterpart σ and σ 2 and need no further motivation. The uual rationale for taking logarithm of the ample tandard deviation i to approximately normalize in order to obtain more accurate reult via tandard tatitical tet for ignificance, with ome mall contant (by convention 1) often firt added to avoid the potential problem of taking a logarithm of zero (uch a due to rounding in data collection). The three mot tandard ignal-to-noie ratio typically are referred to a "nominal-the-bet", "maller-the-better", and "larger-the-better" and have been propoed for the three common ituation for which the experimental objective either are to: achieve repone value a cloe a poible to a deired target value (and with a minimum variability about that target a poible), uch a for a manufacturing dimenion or output voltage, minimize all repone value a much a poible (and with a minimum variability a poible), uch a for percent hrinkage or deviation from round, or maximize all repone value a much a poible (and with a minimum variability a poible), uch a for bond trength or time until failure. 2

3 Alternate Diperion Meaure in DOE Thee three ignal-to-noie ratio uually are etimated, repectively, by: "Nominal-the-bet": S/N ^ N = 10 log y 2 2, "Smaller-the-better": S/N ^ "Larger-the-better": S/N ^ S = -10 log 1 n L = -10 log 1 n n 2 y i i=1 n i=1, and 1. 2 y i A dicued by other, note that an important ditinction of ignal-to-noie ratio i that rather than decoupling the mean and variance into eparate analye, they attempt to form a ingle combined metric of both central tendency and variability (e.g., Gunter (1988) and Hunter (1987)). Alo note that although Taguchi (1986) propoed many other ignal-to-noie ratio (for example, a le common nominal-the-bet ignal-to-noie ratio ha the form S/N N2 = 10 log 2 ), the above three almot excluively tend to be ued in practice. For each ratio and their aociated lo function, the general rationale i to penalize in ome way for deviation from the deired value in term of both location and diperion in a ingle meaure. For further dicuion, ee Box (1988), Hunter (1985, 1987), Phadke (1989), Pignatello and Ramberg (1991), Taguchi (1986), and other. Regardle of which meaure i ued, ignificant factor or interaction typically are identified via the analyi of variance, F tet, t tet, probability plot, dot plot, and the like (ee Box, Hunter, and Hunter (1978), Daniel (1976), and Montgomery (1984)). Unlike the cae of teting for mean effect, however thee tet now are conducted on the 2 k within-run ummary tatitic (in the cae of two-level factorial deign), rather than the n individual outcome within each of the 2 k run, reulting in the lo of 2 k (n - 1) degree of freedom and the need for either at leat one empty column or variance pooling to identify diperion effect. In addition to the above traditional meaure, alo of interet here i whether ome alternative could reult in better variance effect detection. A uggeted earlier, the bet tet criterion for detecting difference in a proce parameter (in thi cae repone variance σ 2 ) may not necearily be baed on the mot familiar direct etimate of that parameter (e.g., the ample variance 2 ). Thi notion i imilar to a tatement by Box (1988) that "the two deiderata - the bet choice of performance meaure and the bet way to employ the data to etimate it - are ditinct and frequently attainable, and they ought not be confued." Following thi reaoning, one general type of alternate meaure might be baed on replacing each of the n oberved repone within a given run i with ome function of that value in uch a manner that each new value now individually provide ome type of meaure of diperion. The primary motivation for thi approach i that each tranformed value now itelf become a type of individual diperion meaure, rather than all within-run i replicate being aggregated into a ingle collective diperion meaure (uch a i, ln( i +1), S/N N, i, et cetera). Statitical analyi then might be conducted on thee individual value, rather than on the traditional ummary meaure, much a one would conduct analyi for mean effect uing 3

4 Alternate Diperion Meaure in DOE each of the n within-run obervation a individual point etimate of the run repone mean µ i. Such an approach would thereby increae the number of aociated degree of freedom, which in turn may reult in tronger detection power, uch a due to a tighter null reference ditribution (although ee below caution). The general tructure of a 2 2 or L 4 replicated deign i hown in Figure 1a, with all of the n repone within each run i replaced in Figure 1b by the correponding abolute deviation of each obervation from the within-run mean, y i,j - y i, where the ubcript i denote a given et of experimental condition (i.e., a run) and the ubcript j denote a given replication of the experiment under thee condition. The abolute deviation value then are analyzed a if they were individual repone uing traditional analyi of variance or other method, analogou to the approach employed in the tudy of central tendency, with the total original number of degree of freedom preerved. Thi may be epecially advantageou for highly Traditional Meaure Experimental Factor Setting Replicate j Reult, y i,j Mean Variance Run (i) A B C (AB) n y i i ln( i +1) S/N N,i y 1,1 y 1,2... y 1,n y 1 1 ln( 1 +1) S/N N, y 2,1 y 2,2... y 2,n y 2 2 ln( 2 +1) S/N N, y 3,1 y 3,2... y 3,n y 3 3 ln( 3 +1) S/N N, y 4,1 y 4,2... y 4,n y 4 4 ln( 4 +1) S/N N,4 Figure 1a: Traditional Structure of Replicated 2 2 Deign Uing Either,, or S/N N a Diperion Meaure Alternate Meaure Experimental Factor Setting Some Function of Replicate Reult Mean Variance Run (i) A B C (AB) n y i y i y i y 1,1 - y 1 y 1,2 - y 1... y 1,n - y 1 y 1 y 1 y y 2,1 - y 2 y 2,2 - y 2... y 2,n - y 2 y 2 y 2 y y 3,1 - y 3 y 3,2 - y 3... y 3,n - y 3 y 3 y 3 y y 4,1 - y 4 y 4,2 - y 4... y 4,n - y 4 y 4 y 4 y 4 Figure 1b: Alternate Structure of Replicated 2 2 Deign Uing y i,j - y i a Diperion Meaure 4

5 Alternate Diperion Meaure in DOE replicated experiment, fairly aturated deign, or mall deign, but alo could produce unpredictable type I error. (Taking abolute value prevent the n within-run (y i,j - y i ) deviation term from mathematically canceling each other out to um to 0.) Comparion of Meaure Approach In order to compare the relative performance of each meaure, everal factor and interaction were included in tandard 2 k-p factorial experimental deign, and the ability to detect different magnitude variance effect of a factor wa examined. Repone value for each run were generated from normal ditribution with parameter pecified by tandard firt-order additive model and µ = β 0 + β r X r r + β r,m X r r m r m X m σ (γ ) = γ 0 + γ r X r + γ X + γ r,m X r X m, r r where X r = Coded etting of factor r, r m r m β r = Mean coefficient (half-effect) of factor r, β r,m = Mean coefficient (half-effect) of interaction X r X m, r m, γ r = Standard deviation coefficient (half-effect) of factor r, r, and γ r,m = Standard deviation coefficient (half-effect) of interaction X r X m, r m, and γ = Standard deviation coefficient (half-effect) of factor S (varied in analyi). Each of the factor etting are determined from the experimental layout and coded a +1 or -1 in the uual manner, with the half-effect ize γ of factor S varied a decribed below. Note that γ repreent the degree to which factor S affect the repone tandard deviation, with γ = 0 indicating that no true effect exit and larger value repreenting larger effect. By iterating acro a range of value for γ, the operating characteritic for each diperion meaure were etimated by filling the experimental array with imluated repone value uing the µ and σ model equation and coded factor etting, calculating each diperion meaure, and conducting analye of variance on thee reult in the uual manner. Thi proce wa repeated 100,000 time at each increment of γ to achieve reaonably accurate reult, with the probability of each meaure ignaling a variance effect for each value of γ etimated a Pr(Signal γ ) = Number of Time F tet wa Significant Total Number of Simulation Replication. 5

6 Alternate Diperion Meaure in DOE A 2 V 5 1 Example A an illutration, in the following analyi four replicate were generated for each of the 16 experimental run of a 2 V 5 1 factorial experiment, with ix hypotheized factor and interaction aigned to column, uing mean and tandard deviation model equation µ = X 1-5X 2 + 7X 3-4X 2 X 3 + 5X 1 X 4 σ = 10 + X X 2 - X 3 + γ 4 X X X 2 X X 1 X 4, uch that the overall repone mean and variance with all factor et at their midpoint are µ Y = 100 and σ Y = 10, repectively (i.e., when all coded term are zero uch that no half-effect are added or ubtracted). The half-effect γ 4 for factor 4 then wa incremented iteratively from γ 4 = 0 to γ 4 = 4 by 0.05 (correponding to half-effect ranging from 0% to 40% of total repone tandard deviation at center point), at each increment of γ 4 repeating the imulation and ubequent analyi of variance (at an α =.05 ignificance level) 100,000 time. Thee reult are hown in Figure 2, with value along the ordinate repreenting the ize of the half-effect of Factor 4 relative to the repone tandard deviation if all X i are et equal to zero, γ 4 /γ 0. Thi caling can be thought of a half the relative reduction in σ poible by changing between X 4 = -1 and X 4 = +1 (auming no active interaction). Alo included in thi analyi i the abolute deviation term raied to the 0.42 power, y i,j - y i.42, a an attempt ^.42 Etimated Probability of Detecting Effect var ^.42 Var Relative Contribution to Standard Deviation at Center Point Figure 2: Relative Performance of Alternate v Traditional Meaure, 2 V 5 1 Example (4 replicate) 6

7 Alternate Diperion Meaure in DOE Probability of Detection Value of γ 4 and γ 4 /γ 0 (Relative Contribution of Factor 4 to Variability) Diperion γ 4 = 0 γ 4 = 1 γ 4 = 2 γ 4 = 3 γ 4 = 4 Meaure (γ 4 /γ 0 = 0) (γ 4 /γ 0 =.1) (γ 4 /γ 0 =.2) (γ 4 /γ 0 =.3) (γ 4 /γ 0 =.4) y j - y y j - y S/N N S/N N S/N S S/N L Table 1: Comparion of Diperion Meaure Performance, 2 V 5 1 Example to normalize the abolute deviation term before forming the tet tatitic a explained below in the Dicuion ection. The ordering from top to bottom of the legend in Figure 2 reflect the relative ordering of each meaure power from bet to wort. Fale alarm probabilitie for each meaure and their power to detect everal effect ize alo are tabulated in Table 1 for further comparion. A thee reult illutrate, the choice of diperion meaure can make a ignificant difference in both the probability of detecting true diperion effect (i.e., power) and the probability of erroneouly ignaling when no effect truly exit (i.e., ignificance). In thi example,,, S/N N, and S/N N2 have comparable operating characteritic and fale alarm probabilitie cloe to the intended α = 0.05 ( α ˆ ), α ˆ , α ˆ S/N(N) , and α ˆ S/N(N2) ). Note that and exhibit lightly higher power for all effect ize, which agree with tudie reported elewhere (Schmidt and Launby (1995), Box (1988), Gunter (1988)). The within-run variance, 2, ha ignificantly lower power, with S/N S and S/N L both eentially being uele with α ˆ S/N(S) and α ˆ S/N(L) and with negligible detection power acro all effect ize. A pointed out by Hunter (1987) and Montgomery (1996), thi i not urpriing given that thee meaure ineparably confound location and diperion effect. Awarene of thee tradeoff can be important information to practitioner in electing a repone meaure and interpreting analyi reult. Interetingly, the abolute deviation meaure exhibit conitently higher power than the other, although at the expene of higher fale alarm probabilitie ( α ˆ = and α ˆ = , repectively), poibly due to the effect of taking abolute value on term lightly le than the average. 7

8 Alternate Diperion Meaure in DOE Adjutment for Equal Fale Alarm Probabilitie Becaue the difference in fale alarm rate noted above make direct comparion difficult, a an exercie the (etimated) ignificance level for each meaure were adjuted empirically to all equal α ˆ = With γ 4 = 0, all 100,000 F tet value for each meaure were orted to locate the empirical 5 th percentile, which then wa ued a an empirical critical value for that meaure, denoted herein a F'. Thee empirical ' Fˆ α =.05 value then were ued in place of the uual critical value a the half-effect γ 4 increaed iteratively a previouly. The reultant α-adjuted operating characteritic for the ame 2 V 5 1 example a above are compared in Figure 3 and Table 2. A hown, the ame performance comparion and ordering a previouly till hold after adjuting all meaure for equal pecificity, but with maller difference in relative power. The conventional meaure,, S/N N, and S/N N2 again are grouped together, with the ample variance 2 exhibiting ignificantly le power, and S/N S and S/N L again being eentially uele for detecting diperion effect. Similar to previouly, both abolute deviation tatitic have greater power than traditional meaure Etimated Probability of Detecting Effect ^.42 var ^.42 Var Relative Contribution to Standard Deviation at Center Point Figure 3: Relative Performance of Alternate v Traditional Meaure, 2 V 5 1 Example (4 replicate) (After empirical adjutment for equal fale alarm probabilitie) Of coure, determining the 5 th empirical percentile of the correponding F value or otherwie adjuting for a deired α probability will not be poible in practice, but the above example illutrate that relative improvement are poible even with fale alarm probabilitie omehow et equal. A illutrated by the following example, imilar benefit alo are 8

9 Alternate Diperion Meaure in DOE Probability of Detection Value of γ 4 and γ 4 /γ 0 (Relative Contribution of Factor 4 to Variability) Diperion γ 4 = 0 γ 4 = 1 γ 4 = 2 γ 4 = 3 γ 4 = 4 Meaure (γ 4 /γ 0 = 0) (γ 4 /γ 0 =.1) (γ 4 /γ 0 =.2) (γ 4 /γ 0 =.3) (γ 4 /γ 0 =.4) y j - y y j - y S/N N S/N N S/N S S/N L Table 2: Comparion of Diperion Meaure Performance, 2 V 5 1 Example (After empirical adjutment for equal fale alarm probabilitie) obtained under other experimental condition (deign ize, number of replicate, degree of aturation). A Second Example (2 3 ) A fairly pare deign wa ued in the above example, reulting in a good number of degree of freedom aociated with the SSE term to etimate between treatment variance (8 for traditional meaure veru 56 for the abolute deviation meaure). Alternatively, if a more aturated 2 3 deign with 5 replicate were ued to tet which of five factor or interaction appear ignificant, the error term would have only 2 degree of freedom uing traditional meaure veru 34 for the abolute deviation meaure. Ue of maller or more aturated deign therefore may impact traditional meaure more adverely than the abolute deviation meaure. In order to examine the relative performance of each meaure in uch cae, 5 replicate for each run of a full factorial 2 3 experimental deign were generated uing the mean and tandard deviation model equation µ = X X 2-1.5X 3-3.5X 1 X X 2 X 3 σ = γ 1 X X X X 1 X X 2 X 3 with the tandard deviation half-effect of factor 1, γ 1, incremented a above and with each analyi of variance again uing α = 0.05 ignificance level. Thee reult are hown in Fig- 9

10 Alternate Diperion Meaure in DOE ure 4a, with Figure 4b again baed on empirically adjuted ˆ α = 0.05 ignificance level. To facilitate further comparion, Table 3a and 3b alo ummarize thee reult ^.42 Etimated Probability of Detecting Effect var ^.42 Var Relative Contribution to Standard Deviation at Center Point Figure 4a: Relative Performance of Alternate v. Traditional Diperion Meaure, 2 3 Example (5 replicate) Etimated Probability of Detecting Effect ^.42 var ^.42 Var Relative Contribution to Standard Deviation at Center Point Figure 4b: Relative Performance of Alternate v. Traditional Diperion Meaure, 2 3 Example (5 replicate) (After empirical adjutment for equal fale alarm probabilitie) 10

11 Alternate Diperion Meaure in DOE Probability of Detection Value of γ 1 and γ 1 /γ 0 (Relative Contribution of Factor 1 to Variability) Diperion γ 1 = 0 γ 1 =.75 γ 1 = 1.5 γ 1 = 2.25 γ 1 = 3 Meaure (γ 1 /γ 0 = 0) (γ 1 /γ 0 =.1) (γ 1 /γ 0 =.2) (γ 1 /γ 0 =.3) (γ 1 /γ 0 =.4 y j - y y j - y S/N N S/N N S/N S S/N L Table 3a: Comparion of Diperion Meaure Performance, 2 3 Example Probability of Detection Value of γ 1 and γ 1 /γ 0 (Relative Contribution of Factor 1 to Variability) Diperion γ 1 = 0 γ 1 =.75 γ 1 = 1.5 γ 1 = 2.25 γ 1 = 3 Meaure (γ 1 /γ 0 = 0) (γ 1 /γ 0 =.1) (γ 1 /γ 0 =.2) (γ 1 /γ 0 =.3) (γ 1 /γ 0 =.4 y j - y y j - y S/N N S/N N S/N S S/N L Table 3b: Comparion of Diperion Meaure Performance, 2 3 Example (After empirical adjutment for equal fale alarm probabilitie) A previouly, note that, S/N N, and S/N N2 exhibit higher power and roughly the ame fale alarm rate a, S/N L and S/N S again are fairly uele, and y i,j - y i and y i,j - y i.42 exhibit higher power acro all magnitude of variance effect. In contrat with the firt example, note that each meaure exhibit lower power a γ 1 increae than previouly, which may 11

12 Alternate Diperion Meaure in DOE be attributable to the difference in deign aturation and degree of freedom aociated with the error term. Ue of the abolute deviation alo may be appealing for another reaon, namely that if the above deign had been fully aturated, tandard ANOVA or other analyi of thee term offer an alternative to the problematic practice of pooling up or down umof-quare that are etimated to have negligible effect. A dicued by Montgomery (1997), "the pooling of mean quare (variance) i a procedure that ha long been known to produce coniderable bia in tet reult." Number of Replicate In order to examine the effect of the number of replication per run on the relative performance of each meaure, Figure 5 and 6 illutrate the effect of uing only n = 2 replicate per run and increaing to 8 replicate per run, repectively, for the ame 2 3 example a above (where previouly n = 5 replicate). Figure 5a and 6a are for the unadjuted cae and Figure 5b and 6b are for the cae with all fale alarm probabilitie empirically adjuted to α ˆ = 0.05 in the manner decribed above. Table 4 alo compare the performance of, y i,j - y i, and y i,j - y i.42 acro a range of other number of replicate. A thee reult illutrate, note that the difference in performance can vary ignificantly for different number of replicate, with the term conitently exhibiting among the bet detection power than other traditional meaure acro all effect and replicate ize. The abolute deviation term exhibit better power but again at the expene of uncontrolled type I error rate. The y i,j - y i fale detection rate for the n = 2 replicate cae i very inflated above the deired α = 0.05 well beyond any ueful point, perhap due to the lack of central limit effect, wherea it appear to decreae to 0 a n increae. A a general rule, therefore, ue of thi meaure for a mall number of replication hould be avoided; mot cae uing four or more replication examined to-date reulted in fale detection rate of roughly 0.10 or lower (with α =.05).. Alo note that none of the meaure are effective in thi example for n = 2. Table 4ummarize the benefit of larger number of replicate for on both power and convergence to the deired α. Dicuion In all examined cae, and both nominal ignal-to-noie ratio conitently produced equal or better power than other traditional meaure, followed by and 2, while the largerthe-better and maller-the-better ratio were relatively uele. The increaed power of y i,j - y i and y i,j - y i.42 to detect true diperion effect i intereting, although at the expene of higher the fale alarm rate. In one 2 V 5 1 example, power to detect diperion half-effect of roughly 25% of the average total repone variance (when all other factor are at their center point) wa increaed to approximately 0.64 from approximately 0.51 for the conventional,, S/N N1, and S/N N2 meaure. Other example ugget imilar benefit for different experiment ize, aturation level, and number of replicate. 12

13 Alternate Diperion Meaure in DOE Etimated Probability of Detecting Effect ^.42 ^.42 var Var Relative Contribution to Standard Deviation at Center Point Figure 5a: 2 Replicate per Run, 2 3 Example 0.3 Etimated Probability of Detecting Effect ^.42 var ^.42 Var Relative Contribution to Standard Deviation at Center Point Figure 5b: 2 Replicate per Run, 2 3 Example (After empirical adjutment for equal fale alarm probabilitie) 13

14 Alternate Diperion Meaure in DOE 1 Etimated Probability of Detecting Effect ^.42 var ^.42 Var Relative Contribution to Standard Deviation at Center Point Figure 6a: 8 Replicate per Run, 2 3 Example 1 ^.42 Etimated Probability of Detecting Effect ^.42 var Var Relative Contribution to Standard Deviation at Center Point Figure 6b: 8 Replicate per Run, 2 3 Example (After empirical adjutment for equal fale alarm probabilitie) 14

15 Alternate Diperion Meaure in DOE Probability of Detection Value of γ 1 and γ 1 /γ 0 (Relative Contribution of Factor 1 to σ) Number of Diperion γ 1 = 0 γ 1 =.75 γ 1 = 1.5 γ 1 = 2.25 γ 1 = 3 Replicate Meaure (γ 1 /γ 0 = 0) (γ 1 /γ 0 =.1) (γ 1 /γ 0 =.2) (γ 1 /γ 0 =.3) (γ 1 /γ 0 = y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y y j - y Table 4: Effect of Number of Replicate (unadjuted 2 3 cae, deired α =.05) Although the abolute deviation meaure are relatively eay to implement by hand or in a preadheet and preent little additional work for the analyt, their fale alarm rate can vary dramatically when uing traditional analyi of variance method. Ideally a more exact mathematical tet baed on the underlying random variable or reference ditribution therefore could be developed. For example, auming Y ~ normal, the abolute deviation from the mean ha a folded or half normal ditribution related to a central chi denity with one degree of freedom (Johnon, Kotz, and Balakrihnan (1994)), uggeting ome type of tet of 15

16 Alternate Diperion Meaure in DOE the upper chi tail. Alternatively, the abolute deviation can be tranformed to approximate normality by raiing it to a power omewhere between 0.35 and 0.55 dependent upon the pecific criteria, with Y - Y.4168 being baed on the Kullback-Leibler information. Reference Box, G. E. P. (1988), Signal-to-Noie Ratio, Performance Criteria, and Tranformation (with dicuion), Technometric, 30, Box, G. E. P., Hunter, W. G., and Hunter, J. S. (1978), Statitic for Experimenter, New York: John Wiley & Son. Byrne, D. M., and Taguchi, S. (1987), The Taguchi Approach to Parameter Deign, Quality Progre, 20(12), Dec 1987, pp Daniel, C. (1976), Application of Statitic to Indutrial Experimentation, New York: John Wiley & Son. Davidian, M. and Carroll R. J. (1987), Variance Function Etimation, Journal of the American Statitical Aociation, 82, Gunter, B. (1988), Dicuion: Signal-to-Noie Ratio, Performance Criteria, and Tranformation, Technometric, 30, Hunter, J. S. (1985), Statitical Deign Applied to Product Deign, Journal of Quality Technology, 17(4), Hunter, J. S. (1987), Signal to Noie Ratio Debated, Quality Progre 20(5), May 1987, pp Johnon, N. L., Kotz, S., and Balakrihnan, N. (1994), Continuou Univariate Ditribution, New York: John Wiley & Son. Montgomery, D. C. (1997), Deign and Analyi of Experiment, 4th ed., New York: John Wiley & Son. Montgomery, D. C. (1996), Introduction to Statitical Quality Control, 3rd ed., New York: John Wiley & Son. Myer, R. H., and Montgomery, D. C. (1995), Repone Surface Methodology, New York: John Wiley & Son. Phadke, M. S. (1989), Quality Engineering Uing Robut Deign, Princeton NJ: Prentice-Hall. Pignatiello, J. J. and Ramberg, J. S. (1991), "The Ten Triumph and Tragedie of Genichi Taguchi, Quality Engineering 4(2), Schmidt, S. R., and Launby, R.G. (1995), Undertanding Indutrial Deigned Experiment, 4th ed., Colorado Spring, CO: Air Academy Pre. Taguchi, G. (1986). Introduction to Quality Engineering, Dearborn, MI: American Supplier Intitute. Vining, G. G. and Schaub, D. (1996), Experimental Deign for Etimating Both Mean and Variance Function, Journal of Quality Technology, 28(2),

Comparing Means: t-tests for Two Independent Samples

Comparing Means: t-tests for Two Independent Samples Comparing ean: t-tet for Two Independent Sample Independent-eaure Deign t-tet for Two Independent Sample Allow reearcher to evaluate the mean difference between two population uing data from two eparate

More information

Lecture 4 Topic 3: General linear models (GLMs), the fundamentals of the analysis of variance (ANOVA), and completely randomized designs (CRDs)

Lecture 4 Topic 3: General linear models (GLMs), the fundamentals of the analysis of variance (ANOVA), and completely randomized designs (CRDs) Lecture 4 Topic 3: General linear model (GLM), the fundamental of the analyi of variance (ANOVA), and completely randomized deign (CRD) The general linear model One population: An obervation i explained

More information

Social Studies 201 Notes for November 14, 2003

Social Studies 201 Notes for November 14, 2003 1 Social Studie 201 Note for November 14, 2003 Etimation of a mean, mall ample ize Section 8.4, p. 501. When a reearcher ha only a mall ample ize available, the central limit theorem doe not apply to the

More information

Source slideplayer.com/fundamentals of Analytical Chemistry, F.J. Holler, S.R.Crouch. Chapter 6: Random Errors in Chemical Analysis

Source slideplayer.com/fundamentals of Analytical Chemistry, F.J. Holler, S.R.Crouch. Chapter 6: Random Errors in Chemical Analysis Source lideplayer.com/fundamental of Analytical Chemitry, F.J. Holler, S.R.Crouch Chapter 6: Random Error in Chemical Analyi Random error are preent in every meaurement no matter how careful the experimenter.

More information

Z a>2 s 1n = X L - m. X L = m + Z a>2 s 1n X L = The decision rule for this one-tail test is

Z a>2 s 1n = X L - m. X L = m + Z a>2 s 1n X L = The decision rule for this one-tail test is M09_BERE8380_12_OM_C09.QD 2/21/11 3:44 PM Page 1 9.6 The Power of a Tet 9.6 The Power of a Tet 1 Section 9.1 defined Type I and Type II error and their aociated rik. Recall that a repreent the probability

More information

Suggested Answers To Exercises. estimates variability in a sampling distribution of random means. About 68% of means fall

Suggested Answers To Exercises. estimates variability in a sampling distribution of random means. About 68% of means fall Beyond Significance Teting ( nd Edition), Rex B. Kline Suggeted Anwer To Exercie Chapter. The tatitic meaure variability among core at the cae level. In a normal ditribution, about 68% of the core fall

More information

Social Studies 201 Notes for March 18, 2005

Social Studies 201 Notes for March 18, 2005 1 Social Studie 201 Note for March 18, 2005 Etimation of a mean, mall ample ize Section 8.4, p. 501. When a reearcher ha only a mall ample ize available, the central limit theorem doe not apply to the

More information

1. The F-test for Equality of Two Variances

1. The F-test for Equality of Two Variances . The F-tet for Equality of Two Variance Previouly we've learned how to tet whether two population mean are equal, uing data from two independent ample. We can alo tet whether two population variance are

More information

Lecture 7: Testing Distributions

Lecture 7: Testing Distributions CSE 5: Sublinear (and Streaming) Algorithm Spring 014 Lecture 7: Teting Ditribution April 1, 014 Lecturer: Paul Beame Scribe: Paul Beame 1 Teting Uniformity of Ditribution We return today to property teting

More information

A Bluffer s Guide to... Sphericity

A Bluffer s Guide to... Sphericity A Bluffer Guide to Sphericity Andy Field Univerity of Suex The ue of repeated meaure, where the ame ubject are teted under a number of condition, ha numerou practical and tatitical benefit. For one thing

More information

Optimal Coordination of Samples in Business Surveys

Optimal Coordination of Samples in Business Surveys Paper preented at the ICES-III, June 8-, 007, Montreal, Quebec, Canada Optimal Coordination of Sample in Buine Survey enka Mach, Ioana Şchiopu-Kratina, Philip T Rei, Jean-Marc Fillion Statitic Canada New

More information

If Y is normally Distributed, then and 2 Y Y 10. σ σ

If Y is normally Distributed, then and 2 Y Y 10. σ σ ull Hypothei Significance Teting V. APS 50 Lecture ote. B. Dudek. ot for General Ditribution. Cla Member Uage Only. Chi-Square and F-Ditribution, and Diperion Tet Recall from Chapter 4 material on: ( )

More information

Jan Purczyński, Kamila Bednarz-Okrzyńska Estimation of the shape parameter of GED distribution for a small sample size

Jan Purczyński, Kamila Bednarz-Okrzyńska Estimation of the shape parameter of GED distribution for a small sample size Jan Purczyńki, Kamila Bednarz-Okrzyńka Etimation of the hape parameter of GED ditribution for a mall ample ize Folia Oeconomica Stetinenia 4()/, 35-46 04 Folia Oeconomica Stetinenia DOI: 0.478/foli-04-003

More information

[Saxena, 2(9): September, 2013] ISSN: Impact Factor: INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY

[Saxena, 2(9): September, 2013] ISSN: Impact Factor: INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY [Saena, (9): September, 0] ISSN: 77-9655 Impact Factor:.85 IJESRT INTERNATIONAL JOURNAL OF ENGINEERING SCIENCES & RESEARCH TECHNOLOGY Contant Stre Accelerated Life Teting Uing Rayleigh Geometric Proce

More information

μ + = σ = D 4 σ = D 3 σ = σ = All units in parts (a) and (b) are in V. (1) x chart: Center = μ = 0.75 UCL =

μ + = σ = D 4 σ = D 3 σ = σ = All units in parts (a) and (b) are in V. (1) x chart: Center = μ = 0.75 UCL = Our online Tutor are available 4*7 to provide Help with Proce control ytem Homework/Aignment or a long term Graduate/Undergraduate Proce control ytem Project. Our Tutor being experienced and proficient

More information

SMALL-SIGNAL STABILITY ASSESSMENT OF THE EUROPEAN POWER SYSTEM BASED ON ADVANCED NEURAL NETWORK METHOD

SMALL-SIGNAL STABILITY ASSESSMENT OF THE EUROPEAN POWER SYSTEM BASED ON ADVANCED NEURAL NETWORK METHOD SMALL-SIGNAL STABILITY ASSESSMENT OF THE EUROPEAN POWER SYSTEM BASED ON ADVANCED NEURAL NETWORK METHOD S.P. Teeuwen, I. Erlich U. Bachmann Univerity of Duiburg, Germany Department of Electrical Power Sytem

More information

Random vs. Deterministic Deployment of Sensors in the Presence of Failures and Placement Errors

Random vs. Deterministic Deployment of Sensors in the Presence of Failures and Placement Errors Random v. Determinitic Deployment of Senor in the Preence of Failure and Placement Error Paul Baliter Univerity of Memphi pbalitr@memphi.edu Santoh Kumar Univerity of Memphi antoh.kumar@memphi.edu Abtract

More information

Asymptotics of ABC. Paul Fearnhead 1, Correspondence: Abstract

Asymptotics of ABC. Paul Fearnhead 1, Correspondence: Abstract Aymptotic of ABC Paul Fearnhead 1, 1 Department of Mathematic and Statitic, Lancater Univerity Correpondence: p.fearnhead@lancater.ac.uk arxiv:1706.07712v1 [tat.me] 23 Jun 2017 Abtract Thi document i due

More information

Stratified Analysis of Probabilities of Causation

Stratified Analysis of Probabilities of Causation Stratified Analyi of Probabilitie of Cauation Manabu Kuroki Sytem Innovation Dept. Oaka Univerity Toyonaka, Oaka, Japan mkuroki@igmath.e.oaka-u.ac.jp Zhihong Cai Biotatitic Dept. Kyoto Univerity Sakyo-ku,

More information

Standard Guide for Conducting Ruggedness Tests 1

Standard Guide for Conducting Ruggedness Tests 1 Deignation: E 69 89 (Reapproved 996) Standard Guide for Conducting Ruggedne Tet AMERICA SOCIETY FOR TESTIG AD MATERIALS 00 Barr Harbor Dr., Wet Conhohocken, PA 948 Reprinted from the Annual Book of ASTM

More information

Chapter 12 Simple Linear Regression

Chapter 12 Simple Linear Regression Chapter 1 Simple Linear Regreion Introduction Exam Score v. Hour Studied Scenario Regreion Analyi ued to quantify the relation between (or more) variable o you can predict the value of one variable baed

More information

SIMPLE LINEAR REGRESSION

SIMPLE LINEAR REGRESSION SIMPLE LINEAR REGRESSION In linear regreion, we conider the frequency ditribution of one variable (Y) at each of everal level of a econd variable (). Y i known a the dependent variable. The variable for

More information

Design spacecraft external surfaces to ensure 95 percent probability of no mission-critical failures from particle impact.

Design spacecraft external surfaces to ensure 95 percent probability of no mission-critical failures from particle impact. PREFERRED RELIABILITY PAGE 1 OF 6 PRACTICES METEOROIDS & SPACE DEBRIS Practice: Deign pacecraft external urface to enure 95 percent probability of no miion-critical failure from particle impact. Benefit:

More information

By Xiaoquan Wen and Matthew Stephens University of Michigan and University of Chicago

By Xiaoquan Wen and Matthew Stephens University of Michigan and University of Chicago Submitted to the Annal of Applied Statitic SUPPLEMENTARY APPENDIX TO BAYESIAN METHODS FOR GENETIC ASSOCIATION ANALYSIS WITH HETEROGENEOUS SUBGROUPS: FROM META-ANALYSES TO GENE-ENVIRONMENT INTERACTIONS

More information

7.2 INVERSE TRANSFORMS AND TRANSFORMS OF DERIVATIVES 281

7.2 INVERSE TRANSFORMS AND TRANSFORMS OF DERIVATIVES 281 72 INVERSE TRANSFORMS AND TRANSFORMS OF DERIVATIVES 28 and i 2 Show how Euler formula (page 33) can then be ued to deduce the reult a ( a) 2 b 2 {e at co bt} {e at in bt} b ( a) 2 b 2 5 Under what condition

More information

Suggestions - Problem Set (a) Show the discriminant condition (1) takes the form. ln ln, # # R R

Suggestions - Problem Set (a) Show the discriminant condition (1) takes the form. ln ln, # # R R Suggetion - Problem Set 3 4.2 (a) Show the dicriminant condition (1) take the form x D Ð.. Ñ. D.. D. ln ln, a deired. We then replace the quantitie. 3ß D3 by their etimate to get the proper form for thi

More information

Estimating floor acceleration in nonlinear multi-story moment-resisting frames

Estimating floor acceleration in nonlinear multi-story moment-resisting frames Etimating floor acceleration in nonlinear multi-tory moment-reiting frame R. Karami Mohammadi Aitant Profeor, Civil Engineering Department, K.N.Tooi Univerity M. Mohammadi M.Sc. Student, Civil Engineering

More information

RaneNote BESSEL FILTER CROSSOVER

RaneNote BESSEL FILTER CROSSOVER RaneNote BESSEL FILTER CROSSOVER A Beel Filter Croover, and It Relation to Other Croover Beel Function Phae Shift Group Delay Beel, 3dB Down Introduction One of the way that a croover may be contructed

More information

Determination of the local contrast of interference fringe patterns using continuous wavelet transform

Determination of the local contrast of interference fringe patterns using continuous wavelet transform Determination of the local contrat of interference fringe pattern uing continuou wavelet tranform Jong Kwang Hyok, Kim Chol Su Intitute of Optic, Department of Phyic, Kim Il Sung Univerity, Pyongyang,

More information

Preemptive scheduling on a small number of hierarchical machines

Preemptive scheduling on a small number of hierarchical machines Available online at www.ciencedirect.com Information and Computation 06 (008) 60 619 www.elevier.com/locate/ic Preemptive cheduling on a mall number of hierarchical machine György Dóa a, Leah Eptein b,

More information

A BATCH-ARRIVAL QUEUE WITH MULTIPLE SERVERS AND FUZZY PARAMETERS: PARAMETRIC PROGRAMMING APPROACH

A BATCH-ARRIVAL QUEUE WITH MULTIPLE SERVERS AND FUZZY PARAMETERS: PARAMETRIC PROGRAMMING APPROACH Mathematical and Computational Application Vol. 11 No. pp. 181-191 006. Aociation for Scientific Reearch A BATCH-ARRIVA QEE WITH MTIPE SERVERS AND FZZY PARAMETERS: PARAMETRIC PROGRAMMING APPROACH Jau-Chuan

More information

ON THE APPROXIMATION ERROR IN HIGH DIMENSIONAL MODEL REPRESENTATION. Xiaoqun Wang

ON THE APPROXIMATION ERROR IN HIGH DIMENSIONAL MODEL REPRESENTATION. Xiaoqun Wang Proceeding of the 2008 Winter Simulation Conference S. J. Maon, R. R. Hill, L. Mönch, O. Roe, T. Jefferon, J. W. Fowler ed. ON THE APPROXIMATION ERROR IN HIGH DIMENSIONAL MODEL REPRESENTATION Xiaoqun Wang

More information

ARTICLE Overcoming the Winner s Curse: Estimating Penetrance Parameters from Case-Control Data

ARTICLE Overcoming the Winner s Curse: Estimating Penetrance Parameters from Case-Control Data ARTICLE Overcoming the Winner Cure: Etimating Penetrance Parameter from Cae-Control Data Sebatian Zöllner and Jonathan K. Pritchard Genomewide aociation tudie are now a widely ued approach in the earch

More information

MINITAB Stat Lab 3

MINITAB Stat Lab 3 MINITAB Stat 20080 Lab 3. Statitical Inference In the previou lab we explained how to make prediction from a imple linear regreion model and alo examined the relationhip between the repone and predictor

More information

The variance theory of the mirror effect in recognition memory

The variance theory of the mirror effect in recognition memory Pychonomic Bulletin & Review 001, 8 (3), 408-438 The variance theory of the mirror effect in recognition memory SVERKER SIKSTRÖM Stockholm Univerity, Stockholm, Sweden The mirror effect refer to a rather

More information

Proactive Serving Decreases User Delay Exponentially: The Light-tailed Service Time Case

Proactive Serving Decreases User Delay Exponentially: The Light-tailed Service Time Case Proactive Serving Decreae Uer Delay Exponentially: The Light-tailed Service Time Cae Shaoquan Zhang, Longbo Huang, Minghua Chen, and Xin Liu Abtract In online ervice ytem, the delay experienced by uer

More information

Combining allele frequency uncertainty and population substructure corrections in forensic DNA calculations

Combining allele frequency uncertainty and population substructure corrections in forensic DNA calculations Combining allele frequency uncertainty and population ubtructure correction in forenic DNA calculation arxiv:1509.08361v2 [tat.ap] 6 Oct 2015 Robert Cowell Faculty of Actuarial Science and Inurance Ca

More information

THE EXPERIMENTAL PERFORMANCE OF A NONLINEAR DYNAMIC VIBRATION ABSORBER

THE EXPERIMENTAL PERFORMANCE OF A NONLINEAR DYNAMIC VIBRATION ABSORBER Proceeding of IMAC XXXI Conference & Expoition on Structural Dynamic February -4 Garden Grove CA USA THE EXPERIMENTAL PERFORMANCE OF A NONLINEAR DYNAMIC VIBRATION ABSORBER Yung-Sheng Hu Neil S Ferguon

More information

White Rose Research Online URL for this paper: Version: Accepted Version

White Rose Research Online URL for this paper:   Version: Accepted Version Thi i a repoitory copy of Identification of nonlinear ytem with non-peritent excitation uing an iterative forward orthogonal leat quare regreion algorithm. White Roe Reearch Online URL for thi paper: http://eprint.whiteroe.ac.uk/107314/

More information

Implied Historical Federal Reserve Bank Behavior Under Uncertainty

Implied Historical Federal Reserve Bank Behavior Under Uncertainty Proceeding of the 2009 IEEE International Conference on Sytem, Man, and Cybernetic San Antonio, TX, USA - October 2009 Implied Hitorical Federal Reerve Bank Behavior Under Uncertainty Muhittin Yilmaz,

More information

USING NONLINEAR CONTROL ALGORITHMS TO IMPROVE THE QUALITY OF SHAKING TABLE TESTS

USING NONLINEAR CONTROL ALGORITHMS TO IMPROVE THE QUALITY OF SHAKING TABLE TESTS October 12-17, 28, Beijing, China USING NONLINEAR CONTR ALGORITHMS TO IMPROVE THE QUALITY OF SHAKING TABLE TESTS T.Y. Yang 1 and A. Schellenberg 2 1 Pot Doctoral Scholar, Dept. of Civil and Env. Eng.,

More information

( ) ( Statistical Equivalence Testing

( ) ( Statistical Equivalence Testing ( Downloaded via 148.51.3.83 on November 1, 018 at 13:8: (UTC). See http://pub.ac.org/haringguideline for option on how to legitimately hare publihed article. 0 BEYOND Gielle B. Limentani Moira C. Ringo

More information

Lecture 10 Filtering: Applied Concepts

Lecture 10 Filtering: Applied Concepts Lecture Filtering: Applied Concept In the previou two lecture, you have learned about finite-impule-repone (FIR) and infinite-impule-repone (IIR) filter. In thee lecture, we introduced the concept of filtering

More information

Clustering Methods without Given Number of Clusters

Clustering Methods without Given Number of Clusters Clutering Method without Given Number of Cluter Peng Xu, Fei Liu Introduction A we now, mean method i a very effective algorithm of clutering. It mot powerful feature i the calability and implicity. However,

More information

Annex-A: RTTOV9 Cloud validation

Annex-A: RTTOV9 Cloud validation RTTOV-91 Science and Validation Plan Annex-A: RTTOV9 Cloud validation Author O Embury C J Merchant The Univerity of Edinburgh Intitute for Atmo. & Environ. Science Crew Building King Building Edinburgh

More information

Unified Correlation between SPT-N and Shear Wave Velocity for all Soil Types

Unified Correlation between SPT-N and Shear Wave Velocity for all Soil Types 6 th International Conference on Earthquake Geotechnical Engineering 1-4 ovember 15 Chritchurch, ew Zealand Unified Correlation between SPT- and Shear Wave Velocity for all Soil Type C.-C. Tai 1 and T.

More information

III.9. THE HYSTERESIS CYCLE OF FERROELECTRIC SUBSTANCES

III.9. THE HYSTERESIS CYCLE OF FERROELECTRIC SUBSTANCES III.9. THE HYSTERESIS CYCLE OF FERROELECTRIC SBSTANCES. Work purpoe The analyi of the behaviour of a ferroelectric ubtance placed in an eternal electric field; the dependence of the electrical polariation

More information

Testing the Equality of Two Pareto Distributions

Testing the Equality of Two Pareto Distributions Proceeding of the World Congre on Engineering 07 Vol II WCE 07, July 5-7, 07, London, U.K. Teting the Equality of Two Pareto Ditribution Huam A. Bayoud, Member, IAENG Abtract Thi paper propoe an overlapping-baed

More information

Math Skills. Scientific Notation. Uncertainty in Measurements. Appendix A5 SKILLS HANDBOOK

Math Skills. Scientific Notation. Uncertainty in Measurements. Appendix A5 SKILLS HANDBOOK ppendix 5 Scientific Notation It i difficult to work with very large or very mall number when they are written in common decimal notation. Uually it i poible to accommodate uch number by changing the SI

More information

A FUNCTIONAL BAYESIAN METHOD FOR THE SOLUTION OF INVERSE PROBLEMS WITH SPATIO-TEMPORAL PARAMETERS AUTHORS: CORRESPONDENCE: ABSTRACT

A FUNCTIONAL BAYESIAN METHOD FOR THE SOLUTION OF INVERSE PROBLEMS WITH SPATIO-TEMPORAL PARAMETERS AUTHORS: CORRESPONDENCE: ABSTRACT A FUNCTIONAL BAYESIAN METHOD FOR THE SOLUTION OF INVERSE PROBLEMS WITH SPATIO-TEMPORAL PARAMETERS AUTHORS: Zenon Medina-Cetina International Centre for Geohazard / Norwegian Geotechnical Intitute Roger

More information

Multipurpose Small Area Estimation

Multipurpose Small Area Estimation Multipurpoe Small Area Etimation Hukum Chandra Univerity of Southampton, U.K. Ray Chamber Univerity of Wollongong, Autralia Weighting and Small Area Etimation Sample urvey are generally multivariate, in

More information

Gain and Phase Margins Based Delay Dependent Stability Analysis of Two- Area LFC System with Communication Delays

Gain and Phase Margins Based Delay Dependent Stability Analysis of Two- Area LFC System with Communication Delays Gain and Phae Margin Baed Delay Dependent Stability Analyi of Two- Area LFC Sytem with Communication Delay Şahin Sönmez and Saffet Ayaun Department of Electrical Engineering, Niğde Ömer Halidemir Univerity,

More information

Quantifying And Specifying The Dynamic Response Of Flowmeters

Quantifying And Specifying The Dynamic Response Of Flowmeters White Paper Quantifying And Specifying The Dynamic Repone Of Flowmeter DP Flow ABSTRACT The dynamic repone characteritic of flowmeter are often incompletely or incorrectly pecified. Thi i often the reult

More information

Estimation of Current Population Variance in Two Successive Occasions

Estimation of Current Population Variance in Two Successive Occasions ISSN 684-8403 Journal of Statitic Volume 7, 00, pp. 54-65 Etimation of Current Population Variance in Two Succeive Occaion Abtract Muhammad Azam, Qamruz Zaman, Salahuddin 3 and Javed Shabbir 4 The problem

More information

Finite Element Analysis of a Fiber Bragg Grating Accelerometer for Performance Optimization

Finite Element Analysis of a Fiber Bragg Grating Accelerometer for Performance Optimization Finite Element Analyi of a Fiber Bragg Grating Accelerometer for Performance Optimization N. Baumallick*, P. Biwa, K. Dagupta and S. Bandyopadhyay Fiber Optic Laboratory, Central Gla and Ceramic Reearch

More information

Acceptance sampling uses sampling procedure to determine whether to

Acceptance sampling uses sampling procedure to determine whether to DOI: 0.545/mji.203.20 Bayeian Repetitive Deferred Sampling Plan Indexed Through Relative Slope K.K. Sureh, S. Umamahewari and K. Pradeepa Veerakumari Department of Statitic, Bharathiar Univerity, Coimbatore,

More information

CHAPTER 8 OBSERVER BASED REDUCED ORDER CONTROLLER DESIGN FOR LARGE SCALE LINEAR DISCRETE-TIME CONTROL SYSTEMS

CHAPTER 8 OBSERVER BASED REDUCED ORDER CONTROLLER DESIGN FOR LARGE SCALE LINEAR DISCRETE-TIME CONTROL SYSTEMS CHAPTER 8 OBSERVER BASED REDUCED ORDER CONTROLLER DESIGN FOR LARGE SCALE LINEAR DISCRETE-TIME CONTROL SYSTEMS 8.1 INTRODUCTION 8.2 REDUCED ORDER MODEL DESIGN FOR LINEAR DISCRETE-TIME CONTROL SYSTEMS 8.3

More information

ASSESSING EXPECTED ACCURACY OF PROBE VEHICLE TRAVEL TIME REPORTS

ASSESSING EXPECTED ACCURACY OF PROBE VEHICLE TRAVEL TIME REPORTS ASSESSING EXPECTED ACCURACY OF PROBE VEHICLE TRAVEL TIME REPORTS By Bruce Hellinga, 1 P.E., and Liping Fu 2 (Reviewed by the Urban Tranportation Diviion) ABSTRACT: The ue of probe vehicle to provide etimate

More information

Evolutionary Algorithms Based Fixed Order Robust Controller Design and Robustness Performance Analysis

Evolutionary Algorithms Based Fixed Order Robust Controller Design and Robustness Performance Analysis Proceeding of 01 4th International Conference on Machine Learning and Computing IPCSIT vol. 5 (01) (01) IACSIT Pre, Singapore Evolutionary Algorithm Baed Fixed Order Robut Controller Deign and Robutne

More information

Lecture 9: Shor s Algorithm

Lecture 9: Shor s Algorithm Quantum Computation (CMU 8-859BB, Fall 05) Lecture 9: Shor Algorithm October 7, 05 Lecturer: Ryan O Donnell Scribe: Sidhanth Mohanty Overview Let u recall the period finding problem that wa et up a a function

More information

Lecture 8: Period Finding: Simon s Problem over Z N

Lecture 8: Period Finding: Simon s Problem over Z N Quantum Computation (CMU 8-859BB, Fall 205) Lecture 8: Period Finding: Simon Problem over Z October 5, 205 Lecturer: John Wright Scribe: icola Rech Problem A mentioned previouly, period finding i a rephraing

More information

Asymptotic Values and Expansions for the Correlation Between Different Measures of Spread. Anirban DasGupta. Purdue University, West Lafayette, IN

Asymptotic Values and Expansions for the Correlation Between Different Measures of Spread. Anirban DasGupta. Purdue University, West Lafayette, IN Aymptotic Value and Expanion for the Correlation Between Different Meaure of Spread Anirban DaGupta Purdue Univerity, Wet Lafayette, IN L.R. Haff UCSD, La Jolla, CA May 31, 2003 ABSTRACT For iid ample

More information

CHAPTER 4 DESIGN OF STATE FEEDBACK CONTROLLERS AND STATE OBSERVERS USING REDUCED ORDER MODEL

CHAPTER 4 DESIGN OF STATE FEEDBACK CONTROLLERS AND STATE OBSERVERS USING REDUCED ORDER MODEL 98 CHAPTER DESIGN OF STATE FEEDBACK CONTROLLERS AND STATE OBSERVERS USING REDUCED ORDER MODEL INTRODUCTION The deign of ytem uing tate pace model for the deign i called a modern control deign and it i

More information

Statistics and Data Analysis

Statistics and Data Analysis Simulation of Propenity Scoring Method Dee H. Wu, Ph.D, David M. Thompon, Ph.D., David Bard, Ph.D. Univerity of Oklahoma Health Science Center, Oklahoma City, OK ABSTRACT In certain clinical trial or obervational

More information

S_LOOP: SINGLE-LOOP FEEDBACK CONTROL SYSTEM ANALYSIS

S_LOOP: SINGLE-LOOP FEEDBACK CONTROL SYSTEM ANALYSIS S_LOOP: SINGLE-LOOP FEEDBACK CONTROL SYSTEM ANALYSIS by Michelle Gretzinger, Daniel Zyngier and Thoma Marlin INTRODUCTION One of the challenge to the engineer learning proce control i relating theoretical

More information

Transitional behaviors in well-graded coarse granular soils. Associate professor, State Key Laboratory of Coal Mine Disaster Dynamics and Control,

Transitional behaviors in well-graded coarse granular soils. Associate professor, State Key Laboratory of Coal Mine Disaster Dynamics and Control, 1 2 Tranitional behavior in well-graded coare granular oil 3 4 Yang Xiao, S.M.ASCE 1, M. R. Coop 2, Hong Liu 3, Hanlong Liu 4 and Jinghan Jiang 5 5 6 7 8 9 1 11 12 13 14 15 16 17 18 19 2 21 22 1. Yang

More information

Chapter 2 Sampling and Quantization. In order to investigate sampling and quantization, the difference between analog

Chapter 2 Sampling and Quantization. In order to investigate sampling and quantization, the difference between analog Chapter Sampling and Quantization.1 Analog and Digital Signal In order to invetigate ampling and quantization, the difference between analog and digital ignal mut be undertood. Analog ignal conit of continuou

More information

Research Article Reliability of Foundation Pile Based on Settlement and a Parameter Sensitivity Analysis

Research Article Reliability of Foundation Pile Based on Settlement and a Parameter Sensitivity Analysis Mathematical Problem in Engineering Volume 2016, Article ID 1659549, 7 page http://dxdoiorg/101155/2016/1659549 Reearch Article Reliability of Foundation Pile Baed on Settlement and a Parameter Senitivity

More information

SIMPLIFIED MODEL FOR EPICYCLIC GEAR INERTIAL CHARACTERISTICS

SIMPLIFIED MODEL FOR EPICYCLIC GEAR INERTIAL CHARACTERISTICS UNIVERSITY OF PITESTI SCIENTIFIC BULLETIN FACULTY OF ECHANICS AND TECHNOLOGY AUTOOTIVE erie, year XVII, no. ( 3 ) SIPLIFIED ODEL FOR EPICYCLIC GEAR INERTIAL CHARACTERISTICS Ciobotaru, Ticuşor *, Feraru,

More information

The Use of MDL to Select among Computational Models of Cognition

The Use of MDL to Select among Computational Models of Cognition The Ue of DL to Select among Computational odel of Cognition In J. yung, ark A. Pitt & Shaobo Zhang Vijay Balaubramanian Department of Pychology David Rittenhoue Laboratorie Ohio State Univerity Univerity

More information

Emittance limitations due to collective effects for the TOTEM beams

Emittance limitations due to collective effects for the TOTEM beams LHC Project ote 45 June 0, 004 Elia.Metral@cern.ch Andre.Verdier@cern.ch Emittance limitation due to collective effect for the TOTEM beam E. Métral and A. Verdier, AB-ABP, CER Keyword: TOTEM, collective

More information

Efficient Methods of Doppler Processing for Coexisting Land and Weather Clutter

Efficient Methods of Doppler Processing for Coexisting Land and Weather Clutter Efficient Method of Doppler Proceing for Coexiting Land and Weather Clutter Ça gatay Candan and A Özgür Yılmaz Middle Eat Technical Univerity METU) Ankara, Turkey ccandan@metuedutr, aoyilmaz@metuedutr

More information

Molecular Dynamics Simulations of Nonequilibrium Effects Associated with Thermally Activated Exothermic Reactions

Molecular Dynamics Simulations of Nonequilibrium Effects Associated with Thermally Activated Exothermic Reactions Original Paper orma, 5, 9 7, Molecular Dynamic Simulation of Nonequilibrium Effect ociated with Thermally ctivated Exothermic Reaction Jerzy GORECKI and Joanna Natalia GORECK Intitute of Phyical Chemitry,

More information

New bounds for Morse clusters

New bounds for Morse clusters New bound for More cluter Tamá Vinkó Advanced Concept Team, European Space Agency, ESTEC Keplerlaan 1, 2201 AZ Noordwijk, The Netherland Tama.Vinko@ea.int and Arnold Neumaier Fakultät für Mathematik, Univerität

More information

On the Isomorphism of Fractional Factorial Designs 1

On the Isomorphism of Fractional Factorial Designs 1 journal of complexity 17, 8697 (2001) doi:10.1006jcom.2000.0569, available online at http:www.idealibrary.com on On the Iomorphim of Fractional Factorial Deign 1 Chang-Xing Ma Department of Statitic, Nankai

More information

NON-GAUSSIAN ERROR DISTRIBUTIONS OF LMC DISTANCE MODULI MEASUREMENTS

NON-GAUSSIAN ERROR DISTRIBUTIONS OF LMC DISTANCE MODULI MEASUREMENTS The Atrophyical Journal, 85:87 (0pp), 05 December 0 05. The American Atronomical Society. All right reerved. doi:0.088/0004-637x/85//87 NON-GAUSSIAN ERROR DISTRIBUTIONS OF LMC DISTANCE MODULI MEASUREMENTS

More information

Regression. What is regression? Linear Regression. Cal State Northridge Ψ320 Andrew Ainsworth PhD

Regression. What is regression? Linear Regression. Cal State Northridge Ψ320 Andrew Ainsworth PhD Regreion Cal State Northridge Ψ30 Andrew Ainworth PhD What i regreion? How do we predict one variable from another? How doe one variable change a the other change? Caue and effect Linear Regreion A technique

More information

After the invention of the steam engine in the late 1700s by the Scottish engineer

After the invention of the steam engine in the late 1700s by the Scottish engineer Introduction to Statitic 22 After the invention of the team engine in the late 1700 by the Scottih engineer Jame Watt, the production of machine-made good became widepread during the 1800. However, it

More information

Publication V by authors

Publication V by authors Publication Kontantin S. Kotov and Jorma J. Kyyrä. 008. nertion lo and network parameter in the analyi of power filter. n: Proceeding of the 008 Nordic Workhop on Power and ndutrial Electronic (NORPE 008).

More information

Confusion matrices. True / False positives / negatives. INF 4300 Classification III Anne Solberg The agenda today: E.g., testing for cancer

Confusion matrices. True / False positives / negatives. INF 4300 Classification III Anne Solberg The agenda today: E.g., testing for cancer INF 4300 Claification III Anne Solberg 29.10.14 The agenda today: More on etimating claifier accuracy Cure of dimenionality knn-claification K-mean clutering x i feature vector for pixel i i- The cla label

More information

EE 508 Lecture 16. Filter Transformations. Lowpass to Bandpass Lowpass to Highpass Lowpass to Band-reject

EE 508 Lecture 16. Filter Transformations. Lowpass to Bandpass Lowpass to Highpass Lowpass to Band-reject EE 508 Lecture 6 Filter Tranformation Lowpa to Bandpa Lowpa to Highpa Lowpa to Band-reject Review from Lat Time Theorem: If the perimeter variation and contact reitance are neglected, the tandard deviation

More information

An estimation approach for autotuning of event-based PI control systems

An estimation approach for autotuning of event-based PI control systems Acta de la XXXIX Jornada de Automática, Badajoz, 5-7 de Septiembre de 08 An etimation approach for autotuning of event-baed PI control ytem Joé Sánchez Moreno, María Guinaldo Loada, Sebatián Dormido Departamento

More information

STATISTICAL SIGNIFICANCE

STATISTICAL SIGNIFICANCE STATISTICAL SIGNIFICANCE EFFECT SIZE More to life than tatitical ignificance Reporting effect ize Turn out a lot of reearcher do not know what preciely p

More information

EC381/MN308 Probability and Some Statistics. Lecture 7 - Outline. Chapter Cumulative Distribution Function (CDF) Continuous Random Variables

EC381/MN308 Probability and Some Statistics. Lecture 7 - Outline. Chapter Cumulative Distribution Function (CDF) Continuous Random Variables EC38/MN38 Probability and Some Statitic Yanni Pachalidi yannip@bu.edu, http://ionia.bu.edu/ Lecture 7 - Outline. Continuou Random Variable Dept. of Manufacturing Engineering Dept. of Electrical and Computer

More information

A Constraint Propagation Algorithm for Determining the Stability Margin. The paper addresses the stability margin assessment for linear systems

A Constraint Propagation Algorithm for Determining the Stability Margin. The paper addresses the stability margin assessment for linear systems A Contraint Propagation Algorithm for Determining the Stability Margin of Linear Parameter Circuit and Sytem Lubomir Kolev and Simona Filipova-Petrakieva Abtract The paper addree the tability margin aement

More information

The Influence of the Load Condition upon the Radial Distribution of Electromagnetic Vibration and Noise in a Three-Phase Squirrel-Cage Induction Motor

The Influence of the Load Condition upon the Radial Distribution of Electromagnetic Vibration and Noise in a Three-Phase Squirrel-Cage Induction Motor The Influence of the Load Condition upon the Radial Ditribution of Electromagnetic Vibration and Noie in a Three-Phae Squirrel-Cage Induction Motor Yuta Sato 1, Iao Hirotuka 1, Kazuo Tuboi 1, Maanori Nakamura

More information

Statistical Downscaling Prediction of Sea Surface Winds over the Global Ocean

Statistical Downscaling Prediction of Sea Surface Winds over the Global Ocean 7938 J O U R N A L O F C L I M A T E VOLUME 26 Statitical Downcaling Prediction of Sea Surface Wind over the Global Ocean CANGJIE SUN AND ADAM H. MONAHAN School of Earth and Ocean Science, Univerity of

More information

CHAPTER 6. Estimation

CHAPTER 6. Estimation CHAPTER 6 Etimation Definition. Statitical inference i the procedure by which we reach a concluion about a population on the bai of information contained in a ample drawn from that population. Definition.

More information

A Simplified Methodology for the Synthesis of Adaptive Flight Control Systems

A Simplified Methodology for the Synthesis of Adaptive Flight Control Systems A Simplified Methodology for the Synthei of Adaptive Flight Control Sytem J.ROUSHANIAN, F.NADJAFI Department of Mechanical Engineering KNT Univerity of Technology 3Mirdamad St. Tehran IRAN Abtract- A implified

More information

Streaming Calculations using the Point-Kernel Code RANKERN

Streaming Calculations using the Point-Kernel Code RANKERN Streaming Calculation uing the Point-Kernel Code RANKERN Steve CHUCAS, Ian CURL AEA Technology, Winfrith Technology Centre, Dorcheter, Doret DT2 8DH, UK RANKERN olve the gamma-ray tranport equation in

More information

Why ANOVA? Analysis of Variance (ANOVA) One-Way ANOVA F-Test. One-Way ANOVA F-Test. One-Way ANOVA F-Test. Completely Randomized Design

Why ANOVA? Analysis of Variance (ANOVA) One-Way ANOVA F-Test. One-Way ANOVA F-Test. One-Way ANOVA F-Test. Completely Randomized Design Why? () Eample: Heart performance core for 3 group of ubject, Non-moer, Moderate moer, 3Heavy moer 3 Comparing More Than Mean.90..0.9.0.00.89.0.99.9.9.98.88.0.0 Average.90.0.00 When comparing three independent

More information

Journal of Econometrics

Journal of Econometrics Journal of Econometric 74 (23) 66 8 Content lit available at SciVere ScienceDirect Journal of Econometric journal homepage: www.elevier.com/locate/jeconom Low-frequency robut cointegration teting Ulrich

More information

Unified Design Method for Flexure and Debonding in FRP Retrofitted RC Beams

Unified Design Method for Flexure and Debonding in FRP Retrofitted RC Beams Unified Deign Method for Flexure and Debonding in FRP Retrofitted RC Beam G.X. Guan, Ph.D. 1 ; and C.J. Burgoyne 2 Abtract Flexural retrofitting of reinforced concrete (RC) beam uing fibre reinforced polymer

More information

Week 3 Statistics for bioinformatics and escience

Week 3 Statistics for bioinformatics and escience Week 3 Statitic for bioinformatic and escience Line Skotte 28. november 2008 2.9.3-4) In thi eercie we conider microrna data from Human and Moue. The data et repreent 685 independent realiation of the

More information

Target-Hardening Decisions Based on Uncertain Multiattribute Terrorist Utility

Target-Hardening Decisions Based on Uncertain Multiattribute Terrorist Utility CREATE Reearch Archive Publihed Article & Paper 1-1-211 Target-Hardening Deciion Baed on Uncertain Multiattribute Terrorit Utility Chen Wang Univerity of Wiconin - Madion, cwang37@wic.edu Vicki M. Bier

More information

NCAAPMT Calculus Challenge Challenge #3 Due: October 26, 2011

NCAAPMT Calculus Challenge Challenge #3 Due: October 26, 2011 NCAAPMT Calculu Challenge 011 01 Challenge #3 Due: October 6, 011 A Model of Traffic Flow Everyone ha at ome time been on a multi-lane highway and encountered road contruction that required the traffic

More information

A NEW LOAD MODEL OF THE PEDESTRIANS LATERAL ACTION

A NEW LOAD MODEL OF THE PEDESTRIANS LATERAL ACTION A NEW LOAD MODEL OF THE PEDESTRIANS LATERAL ACTION Fiammetta VENUTI PhD Politecnico di Torino Torino, IT Luca Bruno Aociate Profeor Politecnico di Torino Torino, IT Summary Thi paper propoe a new load

More information

APPLICATION OF THE SINGLE IMPACT MICROINDENTATION FOR NON- DESTRUCTIVE TESTING OF THE FRACTURE TOUGHNESS OF NONMETALLIC AND POLYMERIC MATERIALS

APPLICATION OF THE SINGLE IMPACT MICROINDENTATION FOR NON- DESTRUCTIVE TESTING OF THE FRACTURE TOUGHNESS OF NONMETALLIC AND POLYMERIC MATERIALS APPLICATION OF THE SINGLE IMPACT MICROINDENTATION FOR NON- DESTRUCTIVE TESTING OF THE FRACTURE TOUGHNESS OF NONMETALLIC AND POLYMERIC MATERIALS REN A. P. INSTITUTE OF APPLIED PHYSICS OF THE NATIONAL ACADEMY

More information

DYNAMIC MODELS FOR CONTROLLER DESIGN

DYNAMIC MODELS FOR CONTROLLER DESIGN DYNAMIC MODELS FOR CONTROLLER DESIGN M.T. Tham (996,999) Dept. of Chemical and Proce Engineering Newcatle upon Tyne, NE 7RU, UK.. INTRODUCTION The problem of deigning a good control ytem i baically that

More information

SOLVING THE KONDO PROBLEM FOR COMPLEX MESOSCOPIC SYSTEMS

SOLVING THE KONDO PROBLEM FOR COMPLEX MESOSCOPIC SYSTEMS SOLVING THE KONDO POBLEM FO COMPLEX MESOSCOPIC SYSTEMS V. DINU and M. ÞOLEA National Intitute of Material Phyic, Bucharet-Magurele P.O. Box MG-7, omania eceived February 21, 2005 Firt we preent the calculation

More information