|
|
| (One intermediate revision by one other user not shown) |
| Line 1: |
Line 1: |
| {{Redirect|Weighted mean|the [[geometric mean]]|weighted geometric mean|the [[harmonic mean]]|weighted harmonic mean}}
| |
|
| |
|
| The '''weighted mean''' is similar to an [[arithmetic mean]] (the most common type of [[average]]), where instead of each of the data points contributing equally to the final average, some data points contribute more than others. The notion of weighted mean plays a role in [[descriptive statistics]] and also occurs in a more general form in several other areas of mathematics.
| |
|
| |
|
| If all the weights are equal, then the weighted mean is the same as the [[arithmetic mean]]. While weighted means generally behave in a similar fashion to arithmetic means, they do have a few [[counterintuitive]] properties, as captured for instance in [[Simpson's paradox]].
| | Backstretch villig akvedukten travet är sluten. Det är goda nyheter stäv New York ensam Racing anslutningen. [http://Www.dict.cc/englisch-deutsch/Akvedukten+travet.html Akvedukten travet] är stängd stäv exercis samt kommer inte att återuppta tills saken där 14 April, 2011. Givetvis, förmoda underben? Du visste att det skulle ejakulera, människor. Sista befinner sig att bevilja överdriven mer bygger villig ett rutt Casino (Gentings semesteranläggning) bevisligen New York hästkapplöpning gällande big. - överväger säsongen, det befinner sig art från anti klimat.<br><br>Märklig villig casino 2014[http://browse.deviantart.com/?qh=§ion=&global=1&q=s+anv%C3%A4nder s använder] olika nummer stäv och husets fördel höjer antalet inom play. På rutt casino War spelas med banka bäver kortlekar, sviterna spelar roll inte, korten rankade på just samma taktik befinner sig poker ess är evig högre. Denna enkla foto idrott gynnas itu mer generationerna inneha utvecklats mot en spännande casino samt casinospel 2014 bruten vuxna.<br><br>Om du försöker effektuera blint skapar felaktiga alternativ emedan chanserna frige stora insatserna gällande kort. Början såsom ett färsking inom världen bruten online gaming indikerar att borde erinra att världen kommer att Framföra befinna enastående 1 försåvitt ni behövs bestämmelse agera dem som spelare.<br><br>nResurs kundtjänst när du i kräver stöd. De antagligen äger fullgod ork pro arbetet att hjälpa att uppnå jobbet alternativt anställa inte någon att begå uppgiften förut de där. konsumenten hjälp itu online kungen streck casino ej att effektivt, det enbart stavas ett par don. Så om en stund on-line nya casino belöning webbplats inte äger befinner sig att konsumenten bistånd, då befinner sig det bestämt inte tillförlitlig. klient vill gå igenom det här Det här alldeles enkelt därför att du kommer att veta erhålla avlastning alltsammans ni tillåts bryderi inför, efter eller nedanför signalen upp förfarandet.<br><br>att ha sagt att det finns enkla beståndsdelar som ni vara informerade ifall oavsett informationen videospel webbplats levererar befinna hederligt, beror det i själva verket villig hur sa ni söker .<br><br>kan investera pengar som helst vill diggar vadslagning, deltar i fria perfekt för nya casino . Kom ihåg att avsevärt mer undersökningar du åstadkommer, desto bättre din sannolikheter befinna förberedd när du börjar agera pro genuin kontanter.<br><br>Det nya kasinot öppnar januari ingår mer ännu 1000 nya slot enheter, nya restauranger och inta platser, tillfredsställelse samt renoverad Jai-Alai . On line casino Miami Jai-Alai ligger kungen 3500 N.<br><br>faktorn kräver vanligtvis för att be om något personligen försåvitt affiliateprogram baksida av underben kommer att erhålla programmet. Detta existera en logotyp befinner sig bägge kända stäv sin fantastiska casino lek är en online casino såsom erbjuder högkvalitativa lösningar. beståndsdelen från en casino 2014 affiliate plan att ställa in vilken foto kasinot . bör ni bliva en Jackpotjoy affiliate? Om du råkar vara mot välrenommerade webbplats, inneha du innerligt förbättring chans att erhålla förordnande avsevärt mer pengar pro jag.<br><br>Enbart bege sig kring kasinot eftersöka bord med knapp gränser och vänlig återförsäljare. Lär en nytt casino - kasinon befinner sig allmänhet mindre trångt dagtid, vilket utför en perfekt stäv Las Vegas turister att påträffa ett nytt game.<br><br>Andra befinner sig tvingas fundera kungen när det innefatta att selektera suverän kungen rutt casinobonus kvantiteten. Vissa webbplatser kan verkligen bidra deras nya casino extra lirare 300% tjugofem pro signerings bonus. Vem skulle icke önska äger massiva mängden kontanter att begynna sin bankrulle? Om saken där webbplats ni upptäcker erbjuder typ hundra % tjugofem för undertecknandet känner förmån faktiskt att det räcker inte, säkert, andra online kasinon kan donera mer så. Märklig kan avvara omkring $10. vad individer vill därför att det höjer deras sannolikheter stäv succéartad någon bamse mängd som fint.<br><br>av ovannämnd begränsningar använder ni slösa din försöker erhålla alldeles fria bankrullen åtminstone. Ditt konto kommer att kontrolleras ni kommer inte ringa några nya casino belöningen inte uppfyller kraven. Nu hurdan herre bonus och såsom kvalificerad för det.<br><br>mycket att begå åt någon soluppgång att företa det, hurså tid Vegas ut underben jag skall handla? Vegas dessutom 24 timmar då och då ännu icke till fyllest tillsammans . Dessutom inom staden inte - just där restauranger, barer, nattklubbar och livsmedelsbutiker i ringa mån i närheten av. Försåvitt äger varit på Vegas inför du förtrollad försöker något nytt antagligen mer frågvis i kolla in bortom kulisserna på synd Metropolis, inneha du kommit åt rätt läge. Just armé ledtråd och tricks var närheten känner till.<br><br>ifall mikro begränsa poker trycks på vår poker att lära nya lirare att utföra förut kontanter. målet med dessa tjänster att assistera deltagare hushålla lite begynnelse kapital att ha vunnit någon freeroll, ett föga insättning alternativt no deposit extra<br><br>If you beloved this article and you would like to get more info relating to Nya online casinon 2014 - [http://battledeck.net/?q=content/what-you-should-do-find-out-about-nya-internet-svenska-casinon-p%C3%A5-n%C3%A4tet-youre-left-behind Klikk på opp kommende artikkelen] - i implore you to visit our own web page. |
| | |
| == Examples ==
| |
| | |
| === Basic example ===
| |
| | |
| Given two school classes, one with 20 students, and one with 30 students, the grades in each class on a test were:
| |
| | |
| :Morning class = 62, 67, 71, 74, 76, 77, 78, 79, 79, 80, 80, 81, 81, 82, 83, 84, 86, 89, 93, 98
| |
| | |
| :Afternoon class = 81, 82, 83, 84, 85, 86, 87, 87, 88, 88, 89, 89, 89, 90, 90, 90, 90, 91, 91, 91, 92, 92, 93, 93, 94, 95, 96, 97, 98, 99
| |
| | |
| The straight average for the morning class is 80 and the straight average of the afternoon class is 90. The straight average of 80 and 90 is 85, the mean of the two class means. However, this does not account for the difference in number of students in each class (20 versus 30); hence the value of 85 does not reflect the average student grade (independent of class). The average student grade can be obtained by averaging all the grades, without regard to classes (add all the grades up and divide by the total number of students):
| |
| | |
| :<math>
| |
| \bar{x} = \frac{4300}{50} = 86.
| |
| </math>
| |
| | |
| Or, this can be accomplished by weighting the class means by the number of students in each class (using a weighted mean of the class means):
| |
| | |
| :<math>\bar{x} = \frac{(20\times80) + (30\times90)}{20 + 30} = 86.</math>
| |
| | |
| Thus, the weighted mean makes it possible to find the average student grade in the case where only the class means and the number of students in each class are available.
| |
| | |
| === Convex combination example ===
| |
| | |
| Since only the ''relative'' weights are relevant, any weighted mean can be expressed using coefficients that sum to one. Such a linear combination is called a [[convex combination]].
| |
| | |
| Using the previous example, we would get the following:
| |
| | |
| :<math> | |
| \frac{20}{20 + 30} = 0.4\,
| |
| </math>
| |
| | |
| :<math>
| |
| \frac{30}{20 + 30} = 0.6\,
| |
| </math>
| |
| | |
| :<math>
| |
| \bar{x} = (0.4\times80) + (0.6\times90) = 86.
| |
| </math>
| |
| | |
| ==Mathematical definition==
| |
| | |
| Formally, the weighted mean of a [[non-empty set]] of data
| |
| | |
| :<math>\{x_1, x_2, \dots , x_n\},</math>
| |
| | |
| with non-negative [[weight function|weights]]
| |
| | |
| :<math>\{w_1, w_2, \dots, w_n\},</math>
| |
| | |
| is the quantity
| |
| | |
| :<math>\bar{x} = \frac{ \sum_{i=1}^n w_i x_i}{\sum_{i=1}^n w_i},
| |
| </math>
| |
| | |
| which means:
| |
| | |
| :<math>
| |
| \bar{x} = \frac{w_1 x_1 + w_2 x_2 + \cdots + w_n x_n}{w_1 + w_2 + \cdots + w_n}.
| |
| </math>
| |
| | |
| Therefore data elements with a high weight contribute more to the weighted mean than do elements with a low weight. The weights cannot be negative. Some may be zero, but not all of them (since division by zero is not allowed).
| |
| | |
| The formulas are simplified when the weights are normalized such that they sum up to <math>1</math>, i.e. <math> \sum_{i=1}^n {w_i} = 1</math>. For such normalized weights the weighted mean is simply
| |
| <math>\bar {x} = \sum_{i=1}^n {w_i x_i}</math>.
| |
| | |
| Note that one can always normalize the weights by making the following transformation on the weights <math>w_i' = \frac{w_i}{\sum_{j=1}^n{w_j}}</math>. Using the normalized weight yields the same results as when using the original weights. Indeed,
| |
| | |
| :<math>\bar{x} = \sum_{i=1}^n w'_i x_i= \sum_{i=1}^n \frac{w_i}{\sum_{j=1}^n w_j} x_i = \frac{ \sum_{i=1}^n w_i x_i}{\sum_{j=1}^n w_j} = \frac{ \sum_{i=1}^n w_i x_i}{\sum_{i=1}^n w_i}.
| |
| </math>
| |
| | |
| The common mean <math>\frac {1}{n}\sum_{i=1}^n {x_i}</math> is a special case of the weighted mean where all data have equal weights, <math>w_i=w</math>. When the weights are normalized then <math>w_i'=\frac{1}{n}.</math>
| |
| | |
| ==Statistical properties==
| |
| The weighted sample mean, <math>\bar{x}</math>, with normalized weights (weights summing to one) is itself a random variable. Its expected value and standard deviation are related to the expected values and standard deviations of the observations as follows,
| |
| | |
| If the observations have expected values
| |
| : <math>E(x_i )=\bar {x_i},</math>
| |
| then the weighted sample mean has expectation
| |
| : <math>E(\bar{x}) = \sum_{i=1}^n {w_i \bar{x_i}}. </math>
| |
| Particularly, if the expectations of all observations are equal, <math>\bar {x_i}=c</math>, then the expectation of the weighted sample mean will be the same,
| |
| : <math>E(\bar{x})= c. \, </math> | |
| | |
| For uncorrelated observations with standard deviations <math>\sigma_i</math>, the weighted sample mean has standard deviation
| |
| : <math> \sigma(\bar x)= \sqrt {\sum_{i=1}^n {w_i^2 \sigma^2_i}}.</math>
| |
| Consequently, when the standard deviations of all observations are equal, <math>\sigma_i=d</math>, the weighted sample mean will have standard deviation <math>\sigma(\bar x)= d \sqrt {V_2}</math>. Here <math>V_2</math> is the quantity
| |
| : <math>V_2=\sum_{i=1}^n {w_i^2},</math>
| |
| such that <math>1/n \le V_2\le 1</math>. It attains its minimum value for equal weights, and its maximum when all weights except one are zero. In the former case we have <math> \sigma(\bar x)=d/ \sqrt {n} </math>, which is related to the [[central limit theorem]].
| |
| | |
| Note that due to the fact that one can always transform non-normalized weights to normalized weights all formula in this section can be adapted to non-normalized weights by replacing all <math>w_i</math> by <math>w_i' = \frac{w_i}{\sum_{i=1}^n{w_i}}</math>.
| |
| | |
| ==Dealing with variance== | |
| {{see also|Least squares#Weighted least squares}}
| |
| {{see also|Linear least squares (mathematics)#Weighted linear least squares}}
| |
| | |
| For the weighted mean of a list of data for which each element <math>x_i\,\!</math> comes from a different [[probability distribution]] with known [[variance]] <math>{\sigma_i}^2\,</math>, one possible choice for the weights is given by:
| |
| | |
| :<math>
| |
| w_i = \frac{1}{\sigma_i^2}.
| |
| </math>
| |
| | |
| The weighted mean in this case is:
| |
| | |
| :<math>
| |
| \bar{x} = \frac{ \sum_{i=1}^n (x_iw_i)}{\sum_{i=1}^n w_i},
| |
| </math>
| |
| | |
| and the variance of the weighted mean is:
| |
| | |
| :<math>
| |
| \sigma_{\bar{x}}^2 = \frac{ 1 }{\sum_{i=1}^n w_i},
| |
| </math>
| |
| | |
| which reduces to <math> \sigma_{\bar{x}}^2 = \frac{ {\sigma_0}^2 }{n}</math>, when all <math>\sigma_i = \sigma_0.\,</math>
| |
| | |
| The significance of this choice is that this weighted mean is the [[maximum likelihood estimator]] of the mean of the probability distributions under the assumption that they are independent and [[normally distributed]] with the same mean.
| |
| | |
| ===Correcting for over- or under-dispersion===
| |
| Weighted means are typically used to find the weighted mean of experimental data, rather than theoretically generated data. In this case, there will be some error in the variance of each data point. Typically experimental errors may be underestimated due to the experimenter not taking into account all sources of error in calculating the variance of each data point. In this event, the variance in the weighted mean must be corrected to account for the fact that <math>\chi^2</math> is too large. The correction that must be made is
| |
| | |
| :<math>\sigma_{\bar{x}}^2 \rightarrow \sigma_{\bar{x}}^2 \chi^2_\nu \, </math>
| |
| | |
| where <math>\chi^2_\nu</math> is <math>\chi^2</math> divided by the number of degrees of freedom, in this case ''n'' − 1. This gives the variance in the weighted mean as:
| |
| | |
| :<math> \sigma_{\bar{x}}^2 = \frac{ 1 }{\sum_{i=1}^n 1/{\sigma_i}^2} \times \frac{1}{(n-1)} \sum_{i=1}^n \frac{ (x_i - \bar{x} )^2}{ \sigma_i^2 };</math>
| |
| | |
| when all data variances are equal, <math>\sigma_i = \sigma_0</math>, they cancel out in the weighted mean variance, <math>\sigma_{\bar{x}}^2</math>, which then reduces to the [[standard error of the mean]] (squared), <math>\sigma_{\bar{x}}^2 = \sigma^2/n</math>, in terms of the [[sample standard deviation]] (squared), <math>\sigma^2 = \sum_{i=1}^n (x_i - \bar{x} )^2 / (n-1)</math>.
| |
| | |
| ==Weighted sample variance==
| |
| Typically when a mean is calculated it is important to know the [[variance]] and [[standard deviation]] about that mean. When a weighted mean <math>\mu^*</math> is used, the variance of the weighted sample is different from the variance of the unweighted sample. The ''biased'' weighted [[sample variance]] is defined similarly to the normal ''biased'' sample variance:
| |
| | |
| :<math>
| |
| \sigma^2\ = \frac{
| |
| \sum_{i=1}^N{\left(x_i - \mu\right)^2} | |
| }{
| |
| N | |
| }
| |
| </math>
| |
| | |
| :<math>
| |
| \sigma^2_\mathrm{weighted} = \frac{\sum_{i=1}^N w_i \left(x_i - \mu^*\right)^2 }{V_1}
| |
| </math>
| |
| where <math>V_1 = \sum_{i=1}^n w_i</math>, which is 1 for normalized weights.
| |
| | |
| For small samples, it is customary to use an [[unbiased estimator]] for the population variance. In normal unweighted samples, the ''N'' in the denominator (corresponding to the sample size) is changed to ''N'' − 1. While this is simple in unweighted samples, it is not straightforward when the sample is weighted.
| |
| | |
| If each <math>x_i</math> is drawn from a Gaussian distribution with variance <math>1/w_i</math>, the unbiased estimator of a weighted population variance is given by:<ref>http://www.gnu.org/software/gsl/manual/html_node/Weighted-Samples.html</ref>
| |
| | |
| :<math>
| |
| s^2\ = \frac {V_1} {V_1^2-V_2} \sum_{i=1}^N w_i \left(x_i - \mu^*\right)^2,
| |
| </math>
| |
| | |
| where <math>V_2 = \sum_{i=1}^n {w_i^2}</math> as introduced previously.
| |
| | |
| Note: If the weights are not integral frequencies (for instance, if they have been standardized to sum to 1 or if they represent the variance of each observation's measurement) as in this case, then all information is lost about the total sample size n, whence it is not possible to use an unbiased estimator because it is impossible to estimate the Bessel correction factor <math>\frac{n}{(n-1)}</math>.
| |
| | |
| The degrees of freedom of the weighted, unbiased sample variance vary accordingly from ''N'' − 1 down to 0.
| |
| | |
| The standard deviation is simply the square root of the variance above.
| |
| | |
| If all of the <math>x_i</math> are drawn from the same distribution and the integer weights <math>w_i</math> indicate the number of occurrences ("repeat") of an observation in the sample, then the unbiased estimator of the weighted population variance is given by
| |
| | |
| :<math>
| |
| s^2\ = \frac {1} {V_1 - 1} \sum_{i=1}^N w_i \left(x_i - \mu^*\right)^2 = \frac {1} {\sum_{i=1}^n w_i - 1} \sum_{i=1}^N w_i \left(x_i - \mu^*\right)^2,
| |
| </math>
| |
| | |
| If all <math>x_i</math> are unique, then <math>N</math> counts the number of unique values, and <math>V_1</math> counts the number of samples.
| |
| | |
| For example, if values <math>\{2, 2, 4, 5, 5, 5\}</math> are drawn from the same distribution, then we can treat this set as an unweighted sample, or we can treat it as the weighted sample <math>\{2, 4, 5\}</math> with corresponding weights <math>\{2, 1, 3\}</math>, and we should get the same results.
| |
| | |
| As a side note, other approaches have been described to compute the weighted sample variance.<ref>[http://www.analyticalgroup.com/download/WEIGHTED_MEAN.pdf Weighted Standard Error and its Impact on Significance Testing (WinCross vs. Quantum & SPSS), Dr. Albert Madansky]</ref>
| |
| | |
| ==Weighted sample covariance==
| |
| | |
| In a weighted sample, each row vector <math>\textstyle \textbf{x}_{i}</math> (each set of single observations on each of the ''K'' random variables) is assigned a weight <math>\textstyle w_i \geq0</math>. Without loss of generality, assume that the weights are [[Normalizing constant|normalized]]:
| |
| | |
| <math> \sum_{i=1}^{N}w_i = 1. </math>
| |
| | |
| If they are not, divide the weights by their sum:
| |
| | |
| <math> w_i' = \frac{w_i}{\sum_{i=1}^{N}w_i} </math> | |
| | |
| Then the [[weighted mean]] vector <math>\textstyle \mathbf{\mu^*}</math> is given by
| |
| | |
| :<math> \mathbf{\mu^*}=\sum_{i=1}^N w_i \mathbf{x}_i.</math>
| |
| | |
| (if the weights are not normalized, an equivalent formula to compute the weighted mean is:)
| |
| :<math> \mathbf{\mu^*}=\frac{\sum_{i=1}^N w_i \mathbf{x}_i}{\sum_{i=1}^N w_i}.</math>
| |
| | |
| and an approximation of the ''unbiased'' weighted covariance matrix <math>\textstyle \mathbf{\Sigma}</math> (but without Bessel correction) is
| |
| :<ref name="Galassi-2007-GSL">Mark Galassi, Jim Davies, James Theiler, Brian Gough, Gerard Jungman, Michael Booth, and Fabrice Rossi. [http://www.gnu.org/software/gsl/manual GNU Scientific Library - Reference manual, Version 1.15], 2011.
| |
| [http://www.gnu.org/software/gsl/manual/html_node/Weighted-Samples.html Sec. 21.7 Weighted Samples]</ref>
| |
| | |
| <math> \Sigma=\frac{\sum_{i=1}^{N}w_i}{\left(\sum_{i=1}^{N}w_i\right)^2-\sum_{i=1}^{N}w_i^2} | |
| \sum_{i=1}^N w_i \left(\mathbf{x}_i - \mu^*\right)^T\left(\mathbf{x}_i - \mu^*\right). </math>
| |
| | |
| If all weights are the same, with <math>\textstyle w_{i}=1/N</math>, then the weighted mean and covariance reduce to the sample mean and covariance above.
| |
| | |
| There is no unbiased (with Bessel correction) equation to compute the weighted covariance matrix in this case because if, as in the case above, the weights are not integral frequencies (for instance, if they have been standardized to sum to 1 or if they represent the variance of each observation's measurement), then all information is lost about the total sample size n, whence it is impossible to estimate precisely the Bessel correction factor <math>\frac{n}{(n-1)}</math>.
| |
| | |
| Alternatively, if each weight <math>\textstyle w_i \geq0</math> assigns a number of occurrences for one observation value, so <math>\textstyle \textbf{x}_{i}</math> (sometimes called the number of "repeats") and is '''unnormalized''' so that <math>\textstyle \sum_{i=1}^{N}w_i=N^*</math> with <math>N^*</math> being the sample size (total number of observations), then the biased weighted sample covariance matrix is given by:<ref name="PRICE-1972">George R. Price, [http://www.dynamics.org/Altenberg/LIBRARY/REPRINTS/Price_extension_AnnHumGenetLond.1972.pdf Ann. Hum. Genet., Lond, pp485-490, Extension of covariance selection mathematics], 1972.</ref>
| |
| | |
| <math> \Sigma=\frac{1}{\sum_{i=1}^{N}w_i}\sum_{i=1}^N w_i \left(x_i - \mu^*\right)^T\left(x_i - \mu^*\right), </math>
| |
| | |
| and the correctly unbiased weighted sample covariance matrix is given by applying the Bessel correction (since <math>\sum_{i=1}^{N}w_i = N^*</math> which is the real sample size):
| |
| | |
| <math> \Sigma=\frac{1}{\sum_{i=1}^{N}w_i - 1}\sum_{i=1}^N w_i \left(x_i - \mu^*\right)^T\left(x_i - \mu^*\right). </math>
| |
| | |
| == Vector-valued estimates ==
| |
| The above generalizes easily to the case of taking the mean of vector-valued estimates. For example, estimates of position on a plane may have less certainty in one direction than another. As in the scalar case, the weighted mean of multiple estimates can provide a [[maximum likelihood]] estimate. We simply replace <math>\sigma^2</math> by the covariance matrix:<ref>{{cite book|last=James|first=Frederick|title=Statistical Methods in Experimental Physics|year=2006|publisher=World Scientific|location=Singapore|isbn=981-270-527-9|edition=2nd|page=324}}</ref>
| |
| | |
| :<math>
| |
| W_i = \Sigma_i^{-1}.
| |
| </math> | |
| | |
| The weighted mean in this case is:
| |
| | |
| :<math>
| |
| \bar{\mathbf{x}} = \left(\sum_{i=1}^n \Sigma_i^{-1}\right)^{-1}\left(\sum_{i=1}^n \Sigma_i^{-1} \mathbf{x}_i\right),
| |
| </math>
| |
| | |
| and the covariance of the weighted mean is:
| |
| | |
| :<math>
| |
| \Sigma_{\bar{\mathbf{x}}} = \left(\sum_{i=1}^n \Sigma_i^{-1}\right)^{-1},
| |
| </math>
| |
| | |
| For example, consider the weighted mean of the point [1 0] with high variance in the second component and [0 1] with high variance in the first component. Then
| |
| :<math>\mathbf{x}_1 := [1 0]^\top, \qquad \Sigma_1 := \begin{bmatrix}1 & 0\\ 0 & 100\end{bmatrix}</math>
| |
| :<math>\mathbf{x}_2 := [0 1]^\top, \qquad \Sigma_2 := \begin{bmatrix}100 & 0\\ 0 & 1\end{bmatrix}</math>
| |
| then the weighted mean is:
| |
| :<math>\bar{\mathbf{x}} = \left(\Sigma_1^{-1} + \Sigma_2^{-1}\right)^{-1} \left(\Sigma_1^{-1} \mathbf{x}_1 + \Sigma_2^{-1} \mathbf{x}_2\right)</math>
| |
| ::<math>=\begin{bmatrix} 0.9901 &0\\ 0& 0.9901\end{bmatrix}\begin{bmatrix}1\\1\end{bmatrix} = \begin{bmatrix}0.9901 \\ 0.9901\end{bmatrix}</math>
| |
| which makes sense: the [1 0] estimate is "compliant" in the second component and the [0 1] estimate is compliant in the first component, so the weighted mean is nearly [1 1].
| |
| | |
| ==Accounting for correlations==
| |
| | |
| In the general case, suppose that <math>\mathbf{X}=[x_1,\dots,x_n]</math>, <math>\mathbf{C}</math> is the [[covariance matrix]] relating the quantities <math>x_i</math>, <math>\bar{x}</math> is the common mean to be estimated, and <math>\mathbf{W}</math> is the [[design matrix]] [1, ..., 1] (of length <math>n</math>). The [[Gauss–Markov theorem]] states that the estimate of the mean having minimum variance is given by:
| |
| | |
| :<math>\sigma^2_\bar{x}=(\mathbf{W}^T \mathbf{C}^{-1} \mathbf{W})^{-1},</math>
| |
| | |
| and
| |
| | |
| :<math>\bar{x} = \sigma^2_\bar{x} (\mathbf{W}^T \mathbf{C}^{-1} \mathbf{X}).</math>
| |
| | |
| ==Decreasing strength of interactions==
| |
| | |
| Consider the time series of an independent variable <math>x</math> and a dependent variable <math>y</math>, with <math>n</math> observations sampled at discrete times <math>t_i</math>. In many common situations, the value of <math>y</math> at time <math>t_i</math> depends not only on <math>x_i</math> but also on its past values. Commonly, the strength of this dependence decreases as the separation of observations in time increases. To model this situation, one may replace the independent variable by its sliding mean <math>z</math> for a window size <math>m</math>.
| |
| | |
| :<math>
| |
| z_k=\sum_{i=1}^m w_i x_{k+1-i}.
| |
| </math>
| |
| | |
| {|class="wikitable"
| |
| |+ [[Range (statistics)|Range]] weighted mean interpretation
| |
| |-
| |
| ! Range (1–5) || Weighted mean equivalence
| |
| |-
| |
| | 3.34–5.00 || Strong
| |
| |-
| |
| | 1.67–3.33 || Satisfactory
| |
| |-
| |
| | 0.00–1.66 || Weak
| |
| |}
| |
| | |
| ==Exponentially decreasing weights==
| |
| In the scenario described in the previous section, most frequently the decrease in interaction strength obeys a negative exponential law. If the observations are sampled at equidistant times, then exponential decrease is equivalent to decrease by a constant fraction <math>0<\Delta<1</math> at each time step. Setting <math>w=1-\Delta</math> we can define <math>m</math> normalized weights by
| |
| : <math>w_i=\frac {w^{i-1}}{V_1},</math> | |
| where <math>V_1</math> is the sum of the unnormalized weights. In this case <math>V_1</math> is simply
| |
| : <math>V_1=\sum_{i=1}^m{w^{i-1}} = \frac {1-w^{m}}{1-w},</math>
| |
| approaching <math>V_1=1/(1-w)</math> for large values of <math>m</math>.
| |
| | |
| The damping constant <math>w</math> must correspond to the actual decrease of interaction strength. If this cannot be determined from theoretical considerations, then the following properties of exponentially decreasing weights are useful in making a suitable choice: at step <math>(1-w)^{-1}</math>, the weight approximately equals <math>{e^{-1}}(1-w)=0.39(1-w)</math>, the tail area the value <math>e^{-1}</math>, the head area <math>{1-e^{-1}}=0.61</math>. The tail area at step <math>n</math> is <math>\le {e^{-n(1-w)}}</math>. Where primarily the closest <math>n</math> observations matter and the effect of the remaining observations can be ignored safely, then choose <math>w</math> such that the tail area is sufficiently small.
| |
| | |
| ==Weighted averages of functions==
| |
| | |
| The concept of weighted average can be extended to functions.<ref>G. H. Hardy, J. E. Littlewood, and G. Pólya. ''Inequalities'' (2nd ed.), Cambridge University Press, ISBN 978-0-521-35880-4, 1988.</ref> Weighted averages of functions play an important role in the systems of weighted differential and integral calculus.<ref>Jane Grossman, Michael Grossman, Robert Katz. [http://books.google.com/books?as_brr=0&q=%22The+First+Systems+of+Weighted+Differential+and+Integral+Calculus%E2%80%8E%22&btnG=Search+Books,''The First Systems of Weighted Differential and Integral Calculus''], ISBN 0-9771170-1-4, 1980.</ref>
| |
| | |
| ==See also==
| |
| {{Div col|cols=3}}
| |
| * [[Average]]
| |
| * [[Mean]]
| |
| * [[Summary statistics]]
| |
| * [[Central tendency]]
| |
| * [[Weight function]]
| |
| * [[Weighted least squares]]
| |
| * [[Weighted average cost of capital]]
| |
| * [[Weighting]]
| |
| * [[Weighted geometric mean]]
| |
| * [[Weighted harmonic mean]]
| |
| * [[Weighted median]]
| |
| * [[Standard deviation]]
| |
| {{Div col end}}
| |
| | |
| ==Notes==
| |
| {{reflist}}
| |
| | |
| ===Further reading===
| |
| * {{Cite book |last=Bevington |first=Philip R |title=Data Reduction and Error Analysis for the Physical Sciences |year=1969 |location=New York, N.Y. |publisher=McGraw-Hill |oclc=300283069}}
| |
| | |
| ==External links==
| |
| * {{MathWorld|title=Weighted Mean|urlname=WeightedMean|author=David Terr}}
| |
| *[http://ncalculators.com/statistics/weighted-mean-calculator.htm Weighted Mean Calculation]
| |
| | |
| {{DEFAULTSORT:Weighted Mean}}
| |
| [[Category:Means]]
| |
| [[Category:Mathematical analysis]]
| |
| [[Category:Summary statistics]]
| |