假設(shè)檢驗(yàn)的概念外文翻譯

1、<p>　　畢 業(yè) 設(shè) 計(jì)（論文）</p><p>　　外 文 文 獻(xiàn) 翻 譯</p><p>　　題 目：What is Hypothesis Testing</p><p>　　學(xué) 院： </p><p>　　專業(yè)名稱： </p><p>　　學(xué) 號(hào)： </p

2、><p>　　學(xué)生姓名： </p><p>　　指導(dǎo)教師： </p><p>　　2012 年 2 月 1 日</p><p><b>　　假設(shè)檢驗(yàn)的概念</b></p><p>　　Enrico Borriello</p><p><b>　　1、引

3、言</b></p><p>　　統(tǒng)計(jì)假設(shè)是指關(guān)于總體參數(shù)的一種假設(shè)。這個(gè)假設(shè)可為真也可以為假。假設(shè)檢驗(yàn)則指統(tǒng)計(jì)員運(yùn)用正式程序來檢驗(yàn)該假設(shè)并確定其真假度，檢測為真則接受結(jié)果，反之則拒絕接受</p><p><b>　　2、統(tǒng)計(jì)假設(shè)</b></p><p>　　檢驗(yàn)統(tǒng)計(jì)假設(shè)的最好方法是檢驗(yàn)整體參數(shù)，但由于其可行性較低故很難運(yùn)用到實(shí)際操作

4、中。因此，研究人員經(jīng)常選擇從整體參數(shù)中隨機(jī)抽樣檢測的方法。如果樣本參數(shù)與統(tǒng)計(jì)假設(shè)的結(jié)果不一致，那原假設(shè)便不成立。</p><p>　　這里有2種統(tǒng)計(jì)假設(shè)。 </p><p>　　零假設(shè) 零假設(shè)，表示為，通常是假設(shè)樣本觀測結(jié)果從純粹的機(jī)會(huì)。 </p><p>　　對(duì)立假設(shè) 對(duì)立假設(shè)，表示為 或是，是假設(shè)樣本中由一些非隨機(jī)的原因決定的觀測值的指標(biāo)。</p>

5、<p>　　例如，假設(shè)我們想確定一個(gè)硬幣的投擲問題是否是公平和合理。一個(gè)零假設(shè)則可以為，硬幣的投擲結(jié)果為，一半為正一半為反。對(duì)立假設(shè)則可以設(shè)定為，出現(xiàn)正面和出現(xiàn)反面的次數(shù)大不相同。象征性而言，以上假設(shè)可以有如下表達(dá)方式</p><p>　　H0: P = 0.5</p><p>　　Ha: P ≠ 0.5</p><p>　　假設(shè)我們投擲硬幣50次，結(jié)果

6、為40次正面，10次背面，基于該結(jié)果，原先的零假設(shè)可以被拒絕。因此得出結(jié)論，由實(shí)驗(yàn)證據(jù)表明，扔硬幣的結(jié)果不會(huì)是正反完全一致。</p><p>　　那么原來的零假設(shè)應(yīng)該是：投擲出正面和反面的次數(shù)一致。即一半幾率為正，一半幾率為背。</p><p>　　零假設(shè)能否被“接受”？</p><p>　　一些研究學(xué)者認(rèn)為一個(gè)零假設(shè)檢驗(yàn)無外乎兩種結(jié)果，要么接受零假設(shè)，要么拒絕之。

7、然后更多的統(tǒng)計(jì)學(xué)家對(duì)于“接受”零假設(shè)依然持保留態(tài)度。相反，他們認(rèn)為正確的表述應(yīng)該是，對(duì)于零假設(shè)，要么拒絕，要么拒絕失敗。為什么要區(qū)分“接受”和“拒絕失敗”？因?yàn)椤敖邮堋奔匆馕吨慵僭O(shè)為真，“拒絕失敗”則指原數(shù)據(jù)不足以支持對(duì)立假設(shè)要優(yōu)于零假設(shè)。</p><p><b>　　3、假設(shè)檢驗(yàn)</b></p><p>　　在樣本數(shù)據(jù)的基礎(chǔ)上，統(tǒng)計(jì)員遵照一定程序來檢驗(yàn)是否能夠拒

8、絕零假設(shè)。該程序被稱為假設(shè)檢驗(yàn)，總共有四個(gè)步驟。</p><p>　　1.陳述假設(shè)。該步驟包括陳述零假設(shè)和對(duì)立假設(shè)，兩者在表述過程中互相排斥，即一方若為真，則另一方必須為假。</p><p>　　2.設(shè)計(jì)分析計(jì)劃。分析計(jì)劃描述了樣本數(shù)據(jù)將如何被分析，并用來評(píng)估零假設(shè)。評(píng)估通常會(huì)重點(diǎn)分析一個(gè)單獨(dú)的檢測數(shù)據(jù)</p><p>　　3.分析樣本數(shù)據(jù)。找出被檢測的各項(xiàng)數(shù)值（例

9、如平均數(shù)，比例數(shù)，t-score，z-score等等），此類數(shù)據(jù)在分析計(jì)劃中已被列出。</p><p>　　4.說明結(jié)果。運(yùn)用在分析計(jì)劃中所列出的決策規(guī)則，如果檢測數(shù)據(jù)與原假設(shè)不相符，基于零假設(shè)的定義，則可以拒絕之。</p><p><b>　　4、決策失誤</b></p><p>　　在一個(gè)假設(shè)檢測里，會(huì)出現(xiàn)的錯(cuò)誤主要有兩種</p>

10、;<p>　　錯(cuò)誤種類1，當(dāng)計(jì)算員在零假設(shè)為真的情況下，采用了拒絕假設(shè)。犯該類錯(cuò)誤的可能性被稱為顯著水平。通常也被稱為alpha，以α表示。</p><p>　　錯(cuò)誤種類2，當(dāng)零假設(shè)為假時(shí)，計(jì)算員拒絕零假設(shè)失敗。犯該類錯(cuò)誤的可能性被稱為beta，由字母β表示。不犯該類錯(cuò)誤的可能性則稱之為檢測能力。</p><p>　　本文研究模糊評(píng)判法在教學(xué)管理系統(tǒng)中的學(xué)生評(píng)價(jià)的應(yīng)用，通過對(duì)

11、影響學(xué)生評(píng)價(jià)的各種因素的分析而對(duì)其賦予不同的權(quán)重，利用模糊評(píng)判法對(duì)學(xué)生做出一個(gè)綜合的評(píng)判。由于本人水平有限，再加上本文是針對(duì)特定地區(qū)的個(gè)別教學(xué)管理系統(tǒng)的評(píng)判，在文中的權(quán)重通過專家調(diào)查分析而來，帶有一定的主觀色彩。</p><p><b>　　5、判斷法則</b></p><p>　　分析計(jì)劃的規(guī)則就是分析零假設(shè)域。在實(shí)踐操作中，統(tǒng)計(jì)學(xué)家描述這些決策規(guī)則的方法，就是參考

12、p值或參考該接受區(qū)。 </p><p>　　p值。作為零假設(shè)是測量值的強(qiáng)有力的證據(jù)。假設(shè)檢驗(yàn)統(tǒng)計(jì)等于S值的概率是觀察一個(gè)檢驗(yàn)統(tǒng)計(jì)量作為極端的假設(shè)是正確的，假設(shè)零假設(shè)是成立的。如果值小于平均值，便接受假設(shè)，反之則拒絕假設(shè)。 </p><p>　　接受域。該區(qū)域的區(qū)間是一個(gè)范圍值。如果檢驗(yàn)結(jié)果落在檢驗(yàn)統(tǒng)計(jì)區(qū)域內(nèi)，該零假設(shè)成立。該地區(qū)接受的判斷標(biāo)準(zhǔn)時(shí)，使第一類誤差的值等于平均值。 </p&

13、gt;<p>　　拒絕域的定義是，集值以外的地區(qū)。如果判斷標(biāo)準(zhǔn)不被接受，則零假設(shè)被拒絕。在這種情況下，我們說，假設(shè)被拒絕在α的區(qū)域外。 </p><p>　　這兩種方法是等效的。一些統(tǒng)計(jì)文本用P值法；其他的文本使用該區(qū)域檢驗(yàn)該方法。在具體的實(shí)例中，不同的方法有不同的應(yīng)用 。</p><p>　　What is Hypothesis Testing?</p>&l

14、t;p>　　Enrico Borriello</p><p>　　1、preface </p><p>　　A statistical hypothesis is an assumption about a population parameter. This assumption may or may not be true. Hypothesis testing refers t

15、o the formal procedures used by statisticians to accept or reject statistical hypotheses.</p><p>　　2、Statistical Hypotheses</p><p>　　The best way to determine whether a statistical hypothesis is

16、 true would be to examine the entire population. Since that is often impractical, researchers typically examine a random sample from the population. If sample data are not consistent with the statistical hypothesis, the

17、hypothesis is rejected.</p><p>　　There are two types of statistical hypotheses.</p><p>　　Null hypothesis. The null hypothesis, denoted by H0, is usually the hypothesis that sample observations r

18、esult purely from chance.</p><p>　　Alternative hypothesis. The alternative hypothesis, denoted by H1 or Ha, is the hypothesis that sample observations are influenced by some non-random cause.</p><

19、p>　　Because For example, suppose we wanted to determine whether a coin was fair and balanced. A null hypothesis might be that half the flips would result in Heads and half, in Tails. The alternative hypothesis might b

20、e that the number of Heads and Tails would be very different. Symbolically, these hypotheses would be expressed as</p><p>　　H0: P = 0.5 Ha: P ≠ 0.5</p><p>　　Suppose we flipped the coin 50 ti

21、mes, resulting in 40 Heads and 10 Tails. Given this result, we would be inclined to reject the null hypothesis. We would conclude, based on the evidence, that the coin was probably not fair and balanced.</p><p

22、>　　Can we accept the null hypothesis ? Some researchers say that a hypotheis test can have one of two outcomes: you accept the null hypothesis or you reject the null hypothesis. Many statisticians, however, take issue

23、 with the notion of “accepting the null hypothesis.” Instead, the say: you reject the null hypothesis or you fail to reject the null hypothesis.</p><p>　　Why the distinction between “acceptance” and “failure

24、 to reject?” Acceptance implies that the null hypothesis is true. Failure to reject implies that the data are not sufficiently persuasive for us to prefer the alternative hypothesis over the null hypothesis.</p>&

25、lt;p>　　3、Hypothesis Tests</p><p>　　Statisticians follow a formal process to determine whether to reject a null hypothesis, based on sample data. This process, called hypothesis testing, consists of four s

26、teps.</p><p>　　(1) State the hypotheses. This involves stating the null and alternative hypotheses. The hypotheses are stated in such a way that they are mutually exclusive. That is, if one is true, the othe

27、r must be false.(2) Formulate an analysis plan. The analysis plan describes how to use sample data to evaluate the null hypothesis. The evaluation often focuses around a single test statistic.(3) Analyze sample data. F

28、ind the value of the test statistic (mean score, proportion, t-score, z-score, etc.) des</p><p>　　4、Decision Errors</p><p>　　Two types of errors can result from a hypothesis test.</p><

29、;p>　　Type I error. A Type I error occurs when the researcher rejects a null hypothesis when it is true. The probability of committing a Type I error is called the significance level. This probability is also called a

30、lpha , and is often denoted by α.</p><p>　　Type II error. A Type II error occurs when the researcher fails to reject a null hypothesis that is false. The probability of committing a Type II error is called

31、beta, and is often denoted by β. The probability of not committing a Type II error is called the power of the test.</p><p>　　5、Decision Rules</p><p>　　The analysis plan includes decision rules f

32、or rejecting the null hypothesis. In practice, statisticians describe these decision rules in two ways - with reference to a P-value or with reference to a region of acceptance.</p><p>　　P-value. The strengt

33、h of evidence in support of a null hypothesis is measured by the P-value. Suppose the test statistic is equal to S. The P-value is the probability of observing a test statistic as extreme as S, assuming the null hypothes

34、is is true. If the P-value is less than the significance level, we reject the null hypothesis.</p><p>　　Region of acceptance. The region of acceprance is a range of values. If the test statistic falls within

35、 the region of acceptance, the null hypothesis is not rejected. The region of acceptance is defined so that the chance of making a Type I error is equal to the significance level.</p><p>　　The set of values

36、outside the region of acceptance is called the region of rejection . If the test statistic falls within the region of rejection, the null hypothesis is rejected. In such cases, we say that the hypothesis has been rejecte

37、d at the α level of significance.</p><p>　　These approaches are equivalent. Some statistics texts use the P-value approach; others use the region of acceptance approach. In subsequent lessons, this tutorial