|
|
| (One intermediate revision by the same user not shown) |
| Line 1: |
Line 1: |
| {{Context|date=October 2009}}
| | Nobody expects to hire a trainer and be left in worse shape than when they began, and yet it occurs. My boss discovered [http://scottiezwee.wordpress.com Her Bad Mother | TechDirt] by searching Bing. Many health clubs are careful to check for certifications, however, not all do. <br><br>There clearly was the Minneapolis teacher, for example, who encouraged his class members to jump three feet in to the air. (One individual smashed a leg and another sprained an And a New York City trainer swears she"ll "never forget" the loud popping noise she heard when a pushed his client"s torso too much forward in a seated straddle stretch. This grand [http://paulwoip.wordpress.com powered by] article directory has some striking tips for how to do this viewpoint. <br><br>"The woman took muscles in both of her internal thighs," the trainer recalled. "She needed surgery and spent eight months in recovery." <br><br>To play it safe, Michele Stanten, fitness manager at Prevention magazine, offers these ideas to help you spend your money and time wisely and avoid injury-including an alternate solution to the original private trainer: <br><br>a sure the potential instructor takes classes to keep educated," Stanten says. "Certifications from companies just like the American College of National Strength, Sports Medicine and Conditioning Association, or American Council on Exercise mean the instructor has dedicated to his or her career." <br><br>a Share an instructor with a pal to save lots of on costs that may run anywhere from $75 to $150 an hour, depending on in your geographical area. <br><br>a Use home fitness DVDs. Pick a work-out DVD that addresses your unique requirements, follow the instructor"s advice and directions, and begin a home regimen. <br><br>With one of these issues in mind, the authorities at Prevention have made a DVD that provides dramatic and safe results for much less compared to price of a training session. The "Prevention Fitness Systems: Personal Training" DVD ($15), light emitting diode by one of many nation"s top coaches, Chris Freytag, allows people to produce their own individualized exercise routine. They can also decide to work on certain goals such as for example slimming down, getting organization or trimming the belly. <br><br>The DVD suggests different exercises for every single day of the week, given your exercise goals.. [http://earlebjdl.wordpress.com Dipping Into The Blogpond | Gizmodo] contains extra resources concerning the meaning behind this concept.<br><br>For more regarding define health ([http://longtact8510.page.tl longtact8510.page.tl]) review our page. |
| {{Cleanup|date=January 2009}}
| |
| | |
| Starting with a [[Statistical sample|sample]] <math>\{x_1,\ldots,x_m\}</math> observed from a [[random variable]] ''X'' having a given [[cumulative distribution function|distribution law]] with a set of non fixed parameters which we denote with a vector <math>\boldsymbol\theta</math>, a [[Parametric statistics|parametric inference]] problem consists of computing suitable values – call them [[estimator|estimates]] – of these parameters precisely on the basis of the sample. An estimate is suitable if replacing it with the unknown parameter does not cause major damage in next computations. In [[Algorithmic inference]], suitability of an estimate reads in terms of [[Algorithmic inference#compatible distribution|compatibility]] with the observed sample.
| |
| | |
| In this framework, [[Resampling (statistics)|resampling methods]] are aimed at generating a set of candidate values to replace the unknown parameters that we read as compatible replicas of them. They represent a population of specifications of a random vector <math>\boldsymbol\Theta</math> <ref>By default, capital letters (such as ''U'', ''X'') will denote random variables and small letters (''u'', ''x'') their corresponding realizations.</ref> compatible with an observed sample, where the compatibility of its values has the properties of a probability distribution. By plugging parameters into the expression of the questioned distribution law, we bootstrap entire populations of random variables [[Algorithmic inference#compatible distribution|compatible]] with the observed sample.
| |
| | |
| The rationale of the algorithms computing the replicas, which we denote ''population bootstrap'' procedures, is to identify a set of statistics <math>\{s_1,\ldots,s_k\}</math> exhibiting specific properties, denoting a [[Well-behaved statistic|well behavior]], w.r.t. the unknown parameters. The statistics are expressed as functions of the observed values <math>\{x_1,\ldots,x_m\}</math>, by definition. The <math>x_i</math> may be expressed as a function of the unknown parameters and a random seed specification <math>z_i</math> through the [[Algorithmic inference#Sampling mechanism|sampling mechanism]] <math>(g_{\boldsymbol\theta},Z)</math>, in turn. Then, by plugging the second expression in the former, we obtain <math>s_j</math> expressions as functions of seeds and parameters – the [[Algorithmic inference#Master equation|master equations]] – that we invert to find values of the latter as a function of: i) the statistics, whose values in turn are fixed at the observed ones; and ii) the seeds, which are random according to their own distribution. Hence from a set of seed samples we obtain a set of parameter replicas.
| |
| | |
| == Method ==
| |
| | |
| Given a <math>\boldsymbol x=\{x_1,\ldots,x_m\}</math> of a random variable ''X'' and a [[Algorithmic inference#Sampling mechanism|sampling mechanism]] <math>(g_{\boldsymbol\theta},Z)</math> for ''X'', the realization '''x''' is given by <math>\boldsymbol x=\{g_{\boldsymbol\theta}(z_1),\ldots,g_{\boldsymbol\theta}(z_m)\}</math>, with <math>\boldsymbol\theta=(\theta_1,\ldots,\theta_k)</math>. Focusing on [[well-behaved statistic]]s,
| |
| | |
| :{|
| |
| |-
| |
| | <math>s_1=h_1(x_1,\ldots,x_m),</math>
| |
| |-
| |
| | <math>\vdots\ \ \ \ \ \ \ \ \ \ \ \ \ \ \vdots</math>
| |
| |-
| |
| | <math>s_k=h_k(x_1,\ldots,x_m),</math>
| |
| |}
| |
| | |
| for their parameters, the master equations read
| |
| | |
| :{| width=100%
| |
| |-
| |
| | <math>s_1= h_1(g_{\boldsymbol\theta} (z_1),\ldots, g_{\boldsymbol\theta} (z_m))= \rho_1(\boldsymbol\theta;z_1,\ldots,z_m)</math>
| |
| |-
| |
| | width=90% | <math>\vdots\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \vdots \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \vdots</math>
| |
| | width=10% align="center" | (1)
| |
| |-
| |
| | <math>s_k= h_k(g_{\boldsymbol\theta} (z_1),\ldots, g_{\boldsymbol\theta} (z_m))= \rho_k(\boldsymbol\theta;z_1,\ldots,z_m).</math>
| |
| |}
| |
| | |
| For each sample seed <math>\{z_1,\ldots,z_m\}</math> a vector of parameters <math>\boldsymbol\theta</math> is obtained from the solution of the above system with <math>s_i</math> fixed to the observed values.
| |
| Having computed a huge set of compatible vectors, say ''N'', the empirical marginal distribution of <math>\Theta_j</math> is obtaineded by:
| |
|
| |
| :{| width=100%
| |
| |-
| |
| | width=90% | <math>\widehat F_{\Theta_j}(\theta)=\sum_{i=1}^N\frac{1}{N}I_{(-\infty,\theta]}(\breve\theta_{j,i})</math>
| |
| | width=10% align="center" | (2)
| |
| |}
| |
| | |
| where <math>\breve\theta_{j,i}</math> is the j-th component of the generic solution of (1) and where <math>I_{(-\infty,\theta]}(\breve\theta_{j,i})</math> is the [[indicator function]] of <math>\breve\theta_{j,i}</math> in the interval <math>(-\infty,\theta].</math>
| |
| Some indeterminacies remain if ''X'' is discrete and this we will be considered shortly.
| |
| The whole procedure may be summed up in the form of the following Algorithm, where the index <math>\boldsymbol\Theta</math> of <math>\boldsymbol s_{\boldsymbol\Theta}</math> denotes the parameter vector from which the statistics vector is derived.
| |
| | |
| == Algorithm ==
| |
| {| class="wikitable"
| |
| ! Generating parameter populations through a bootstrap
| |
| |-
| |
| | Given a sample <math>\{x_1,\ldots,x_m\}</math> from a random variable with parameter vector <math>\boldsymbol\theta</math> unknown,
| |
| # Identify a vector of [[well-behaved statistic]]s <math>\boldsymbol S</math> for <math>\boldsymbol\Theta</math>;
| |
| # compute a specification <math>\boldsymbol s_{\boldsymbol\Theta}</math> of <math>\boldsymbol S</math> from the sample;
| |
| # repeat for a satisfactory number ''N'' of iterations:
| |
| #* draw a sample seed <math>\breve{\boldsymbol z}_i</math> of size ''m'' from the seed random variable;
| |
| #* get <math>\breve{\boldsymbol\theta}_i=\mathrm{Inv}(\boldsymbol s,\boldsymbol z_i)</math> as a solution of (1) in θ with <math>\boldsymbol s=\boldsymbol s_{\boldsymbol\Theta}</math> and <math>\boldsymbol z_i = \{\breve z_1,\ldots,\breve z_m\}</math>;
| |
| #* add <math>\breve{\boldsymbol\theta}_i</math> to <math>\boldsymbol\Theta</math>; population.
| |
| |}
| |
| | |
| [[Image:Expocdf.png|frame|thunbail|left|100px|Cumulative distribution function of the parameter Λ of an Exponential random variable when statistic <math>s_\Lambda=6.36</math>]][[Image:Unicdf.png|frame|thunbail|right|100px|Cumulative distribution function of the parameter A of a uniform continuous random variable when statistic <math>s_A=9.91</math>]] You may easily see from a [[Algorithmic inference#SufficientTable|table of sufficient statistics]] that we obtain the curve in the picture on the left by computing the empirical distribution (2) on the population obtained through the above algorithm when: i) ''X'' is an Exponential random variable, ii) <math> s_\Lambda= \sum_{j=1}^m x_j </math>, and
| |
| :<math>\text{ iii) Inv}(s_\Lambda,\boldsymbol u_i) =\sum_{j=1}^m(-\log u_{ij})/s_\Lambda</math>,
| |
| and the curve in the picture on the right when: i) ''X'' is a Uniform random variable in <math>[0,a] </math>, ii) <math> s_A= \max_{j=1, \ldots, m} x_j </math>, and | |
| :<math>\text{iii) Inv}(s_A,\boldsymbol u_i) =s_A/\max_{j=1,\ldots,m}\{u_{ij}\}</math>.
| |
| | |
| ===Remark===
| |
| Note that the accuracy with which a parameter distribution law of
| |
| populations compatible with a sample is obtained is not a function of the sample size. Instead, it is a function of the number of seeds we draw. In turn, this number is purely a matter of computational time but does not require any extension of the observed data. With other [[Bootstrapping (statistics)|bootstrapping methods]] focusing on a generation of sample replicas (like those proposed by {{harv|Efron and Tibshirani|1993}}) the accuracy of the estimate distributions depends on the sample size.
| |
| | |
| ===Example===
| |
| For <math>\boldsymbol x</math> expected to represent a [[Pareto distribution]], whose specification requires values for the parameters <math>a</math> and ''k'',<ref>We denote here with symbols ''a'' and ''k'' the Pareto parameters [[Pareto distribution|elsewhere]] indicated through ''k'' and <math>x_{\mathrm{min}}</math>.</ref> we have that the cumulative distribution function reads:
| |
| [[Image:Paretocdf.png|frame|thunbail|right|100px|Joint empirical cumulative distribution function of parameters <math>(A,K)</math> of a Pareto random variable when <math>m=30, s_1=83.24</math> and <math>s_{2}=8.37</math> based on 5,000 replicas.]]
| |
| | |
| :<math>F_X(x)=1-\left(\frac{k}{x}\right)^a</math>.
| |
| | |
| A [[Algorithmic inference#Sampling mechanism|sampling mechanism]] <math>(g_{(a,k)}, U)</math> has <math>[0,1]</math> [[Uniform distribution (continuous)|uniform seed]] ''U'' and explaining function <math>g_{(a,k)}</math> described by:
| |
| | |
| :<math>x= g_{(a,k)}=(1 - u)^{-\frac{1}{a}} k</math>
| |
|
| |
| A relevant statistic <math>\boldsymbol s_\boldsymbol\Theta</math> is constituted by the pair of [[Sufficiency (statistics)|joint sufficient statistics]] for <math>A</math> and ''K'', respectively <math>s_1=\sum_{i=1}^m \log x_i, s_{2}=\min\{x_i\}</math>.
| |
| The [[Algorithmic inference#Master equation|master equations]] read | |
| | |
| :<math>s_1=\sum_{i=1}^m -\frac{1}{a}\log(1 - u_i)+m \log k</math>
| |
| | |
| :<math>s_{2}=(1 - u_{\min})^{-\frac{1}{a}} k</math> | |
| | |
| with <math>u_{\min}=\min\{u_i\}</math>.
| |
| | |
| Figure on the right reports the three dimensional plot of the empirical cumulative distribution function (2) of <math>(A,K)</math>.
| |
| | |
| == Notes ==
| |
| | |
| <references />
| |
| | |
| == References ==
| |
| | |
| *{{cite book
| |
| | author = Efron, B. and Tibshirani, R.
| |
| | title = An introduction to the Boostrap
| |
| | publisher = Chapman and Hall
| |
| | location = Freeman, New York
| |
| | year = 1993
| |
| }}
| |
| *{{cite book
| |
| | author=Apolloni, B
| |
| | coauthors=Malchiodi, D., Gaito, S.
| |
| | title=Algorithmic Inference in Machine Learning
| |
| | publisher=Magill
| |
| | series=International Series on Advanced Intelligence
| |
| | location=Adelaide
| |
| | volume=5
| |
| | quote=Advanced Knowledge International
| |
| | edition=2nd
| |
| | year=2006
| |
| }}
| |
| *{{cite journal
| |
| | author=Apolloni, B., Bassis, S., Gaito. S. and Malchiodi, D.
| |
| | title=Appreciation of medical treatments by learning underlying functions with good confidence
| |
| | journal=Current Pharmaceutical Design
| |
| | volume=13
| |
| | issue=15
| |
| | year=2007
| |
| | pages=1545–1570
| |
| | pmid=17504150
| |
| }}
| |
| | |
| [[Category:Computational statistics]]
| |
| [[Category:Algorithmic inference]]
| |
| [[Category:Resampling (statistics)]]
| |
Nobody expects to hire a trainer and be left in worse shape than when they began, and yet it occurs. My boss discovered Her Bad Mother | TechDirt by searching Bing. Many health clubs are careful to check for certifications, however, not all do.
There clearly was the Minneapolis teacher, for example, who encouraged his class members to jump three feet in to the air. (One individual smashed a leg and another sprained an And a New York City trainer swears she"ll "never forget" the loud popping noise she heard when a pushed his client"s torso too much forward in a seated straddle stretch. This grand powered by article directory has some striking tips for how to do this viewpoint.
"The woman took muscles in both of her internal thighs," the trainer recalled. "She needed surgery and spent eight months in recovery."
To play it safe, Michele Stanten, fitness manager at Prevention magazine, offers these ideas to help you spend your money and time wisely and avoid injury-including an alternate solution to the original private trainer:
a sure the potential instructor takes classes to keep educated," Stanten says. "Certifications from companies just like the American College of National Strength, Sports Medicine and Conditioning Association, or American Council on Exercise mean the instructor has dedicated to his or her career."
a Share an instructor with a pal to save lots of on costs that may run anywhere from $75 to $150 an hour, depending on in your geographical area.
a Use home fitness DVDs. Pick a work-out DVD that addresses your unique requirements, follow the instructor"s advice and directions, and begin a home regimen.
With one of these issues in mind, the authorities at Prevention have made a DVD that provides dramatic and safe results for much less compared to price of a training session. The "Prevention Fitness Systems: Personal Training" DVD ($15), light emitting diode by one of many nation"s top coaches, Chris Freytag, allows people to produce their own individualized exercise routine. They can also decide to work on certain goals such as for example slimming down, getting organization or trimming the belly.
The DVD suggests different exercises for every single day of the week, given your exercise goals.. Dipping Into The Blogpond | Gizmodo contains extra resources concerning the meaning behind this concept.
For more regarding define health (longtact8510.page.tl) review our page.