有偏樣本 的英文怎麼說
中文拼音 [yǒupiānyàngběn]
有偏樣本
英文
biased sample- 有 : 有副詞[書面語] (表示整數之外再加零數): 30 有 5 thirty-five; 10 有 5年 fifteen years
- 偏 : Ⅰ形容詞1 (不正; 歪斜) inclined to one side; slanting; leaning 2 (只側重一面) partial; prejudi...
- 樣 : Ⅰ名詞1. (形狀) appearance; shape 2. (樣品) sample; model; pattern Ⅱ量詞(表示事物的種類) kind; type
- 本 : i 名詞1 (草木的莖或根)stem or root of plants 2 (事物的根源)foundation; origin; basis 3 (本錢...
- 樣本 : sample book; specimen; advanced copy; sample; muster; scantling; instance; statistics
-
In addition, this paper makes a comparison among the pls, canonical correlation analysis and principal components analysis ( pca ) method, which proves that the pls works better than the others when there are only the small sample and more dependent and independent variables. chen yinghui ( lndustry economics ) directed by : pro
本文在交通運輸業與旅遊業的關系分析中,對偏最小二乘回歸方法與主成分分析方法作了比較,驗證了在樣本較少、多對多的回歸分析中,偏最小二乘方法有明顯的優勢。Compared with the classical bp algorithm, robust adaptive bp algorithm possesses some advantages as following : ( 1 ) increasing the accuracy of the network training by means of using both the relative and absolute residual to adjust the weight values ; ( 2 ) improve the robustness and the network convergence rate through combining with the robust statistic technique by way of judging the values of the samples " relative residual to establish the energy function so that can suppress the effect on network training because of the samples with high noise disturbances ; ( 3 ) prevent entrapping into the local minima area and obtain the global optimal result owing to setting the learning rate to be the function of the errors and the error gradients when network is trained. the learning rate of the weights update change with the error values of the network adaptively so that can easily get rid of the disadvantage of the classical bp algorithm that is liable to entrap into the local minima areas
與基本bp演算法相比,本文提出的魯棒自適應bp演算法具有以下優點: ( 1 )與魯棒統計技術相結合,通過訓練樣本相對偏差的大小,確定不同訓練樣本對能量函數的貢獻,來抑制含高噪聲干擾樣本對網路訓練的不良影響,從而增強訓練的魯棒性,提高網路訓練的收斂速度; ( 2 )採用相對偏差和絕對偏差兩種偏差形式對權值進行調整,提高了網路的訓練精度; ( 3 )在採用梯度下降演算法對權值進行調整的基礎上,通過將學習速率設為訓練誤差及誤差梯度的特殊函數,使學習速率依賴于網路訓練時誤差瞬時的變化而自適應的改變,從而可以克服基本bp演算法容易陷入局部極小區域的弊端,使訓練過程能夠很快的「跳出」局部極小區域而達到全局最優。Through the theory of light radiation and intensity, we can use the fewest leds to satisfy the luminous intensity demand. through image segmentation theory, we can accurately pick module up from the test stripe when it is put in wrong directions. through image processing theory, we can acquire correct information and avoid the bad effects from the asymmetric chemistry reaction and instability of the devices
用光的輻射和強度理論,我們計算出了獲得足夠圖像強度所需的最少光源;用圖像分割理論,我們在試紙條傾斜放置或有垂直方向上的偏移時,準確地提取出了各模塊的數據;用平滑濾波和均值濾波理論,我們濾除了由於反應不均勻及硬體設備不穩定帶來的噪聲;用交遇區設計線性分類器的方法,我們降低了有限樣本設計線性分類器帶來的誤差,提高了檢驗準確度。( 5 ) study of cvm shows that the wtp are on the low side, although the wtp reflects the basic trend that respondents are able to answer the wtp correspond to the changes of ecological environment, if a large amount of samples are collected at the existing condition in our country
運用條件價值法,在現有的條件下,通過大量樣本的被調查者回答的支付意願,可以揭示他們對環境變化做出反應的基本趨勢,但結果偏低。The new measures are a function of the genetic distance between the marker locus and a qtl. through simulations, we found that when marker allele frequencies vary across loci, the previous hwd measures are biased and not powe
計算機模擬表明,當各標記基因頻率不同時,用以前的hwd指數精細定位會產生偏差,新的指數可以有效的進行精細定位,使用y > u和y < t的樣本的lcd的功效普遍比僅僅使用y > u的樣本的氣回,的功效高。Comparing with non - bnyain methods, it ' s prominent featares lay in that it combines the prior and posterior information, which avoids the disadvantag of subjective bias caused by simply using the prior information only, of blind search caused by the incomplete sample information, of noise affection caused by simply using the sample information only if we choice a suitable priof, we can conduct the bayesian leaming effectively, so it fits the problems of data mining and machine leaming that possess charaters of probability and statistics, especially when the samples are rare
與非貝葉揚方法相比,貝葉斯方法的特出特點是其學習機制可以綜合先驗信息和后驗信息,既可避免只使用先驗信息可能帶來的主觀偏見,和缺乏樣本信息時的大量盲目搜索與計算,也可避免只使用樣本信息帶來的噪音的影響只要合理地確定先驗,就可以進行有效的學習。因此,適用於具有概率統計特徵的數據採掘和機器學習(或發現)問題,尤其是樣本難得的問題The differences in orbital morphologies due to different potentials is slighting, however, given a certain potential, for clusters that have perigalactic distance smaller than 1 kpc, some orbits may exhibit a chaotic behavior. the correlation between the metallicity of samples and the orbital morphologies is unclearly ; ( 4 ) it is found that the semi - major axis, apogalactic distance and azimuth period of 29 sample clusters are changed with their metallicity similarly, but a obvious correlation is seen between orbital eccentricity and metallicity. there is a fraction of 24 % of the sample clusters with eccentricities lower than 0. 4
不同的引力勢模型對球狀星團軌道的具體形態影響不大,在給定的引力勢模型下,當某些星團的運動軌道穿越距銀心1kpc附近的區域時會出現「混沌」現象: ( 4 ) 29個樣本星團的軌道半長軸、遠銀心距和方位周期隨金屬度的變化規律基本相似,樣本星團的金屬度與其軌道形態之間的相關性並不明顯,然而軌道偏心率與金屬度有關,對于所選的暈族樣本星團而言,大約有24的樣本星團的軌道偏心率低於0 . 4 。The different galactic gravitational potential have not clear influence upon the perigalactic distance, eccentricity and uncertainty of orbital parameters, but which is significant for other parameters, such as the apogalactic distance, semi - major axis, radial period and azimuth period and so on ; ( 5 ) the monte carlo simulation show that : the distribution morphologies of the orbital parameters which are produced by simulation data are separated into three types : gaussian profile, semi - gaussian profile and non - gaussian profile
不同的引力勢模型對近銀心距、偏心率和參數的不確定度等量影響較小,但是對遠銀心距、軌道半長軸、徑向周期和方位周期等參數影響較為明顯。 ( 5 ) montecarlo模擬的結果表明:根據模擬觀測數據得到的軌道參數的分佈形態大致可分為三類:高斯分佈、準高斯分佈和非高斯分佈;觀測數據對軌道參數的影響與樣本的選取、軌道參數的類型等有關。According to the requirements to pd pattern auto - recognition, this paper studies systematically the basic theories and realizable methods for auto - recognition of pd gray intensity image : ( 1 ) in the requirement of on - line pd monitoring for transformer, several discharge models are designed and the relevant experiment methods projected. with discharge model tests, a lot of discharge sample data is acquired. on the base of systematical research on recognition for pd gray intensity image, this paper puts forward two kinds of fractal features, the 2nd generalized dimensions of original pd images and fractal dimensions of high gray intensity pd images, and then the relevant extraction methods
針對局部放電模式自動識別的需要,作者系統地研究了局部放電灰度圖像自動識別中的基本理論和實現方法: ( 1 )根據變壓器局部放電在線監測的要求,設計了放電模型和實驗方法,並通過模型實驗獲得了大量放電樣本數據,為構造局部放電灰度圖像和採用bpnn進行識別作好準備; ( 2 )研究了局部放電灰度圖像的構造方法以及降維構造32 32灰度和矩陣的方法;在用人工神經網路對局部放電進行模式識別時,分析了bp網路的優缺點,對典型bp網路的結構和學習訓練演算法提出了改進,採用帶有偏差單元的遞歸神經網路作為模式分類器;採用32 32灰度和矩陣進行bpnn識別結果表明這種方法是有效的。This paper uses arch model method in econometrics to set up an auto - regression model with different variance characteristic, which catches to the signal of herd behavior that can be comparatively sensitive. basing on the sample stocks of the index 180 of sse for studying sample, author conduct empirical tests on the non - linear relations between csad ( cross - sectional absolute deviation of returns ) and the market returns to judge whether the herd behavior in the stock market of china is remarkable. according to the empirical analysis, author finds, both in the up - market and down - market, certain herd behavior exist on the stock market of our country
本文運用計量經濟學中的arch模型方法,建立了一個能較為敏感的捕捉到羊群行為信號的具有異方差特性的自回歸模型,以上證180指數樣本股為研究樣本,通過檢驗個股截面收益的絕對偏差( csad )與市場組合收益的非線性關系,來判斷我國股市羊群行為是否顯著,通過實證分析,我們發現,無論是市場上漲階段還是下跌階段,我國股市都存在一定的羊群行為,同時,本文通過比較分析,對實證結果進行深入的剖析,對羊群行為的形成原因進行簡要的分析,並對如何控制羊群行為提出了一些政策性建議。As hurst parameter estimated bias exists, the precision may be improved by using non - linear estimate, where arfima model is proposed and used for verification
從統計結果來看,樣本序列呈現出尖峰、胖尾等有偏特徵,明顯不滿足正態分佈的假設,表明收益序列可能具有長程相關或記憶性。3. the methodology based upon lssvm for on - line inferential measurement has been proposed and applied in parameter estimation problem encountered in a continuous catalytic reforming unit. to deal with model deviation problem, moving windows framework is incorporated into inferential measurement method that is based upon lssvm
將svm運用於連續重整裝置中的參數預估問題,在簡要介紹及說明了工程應用中存在的問題之後,如訓練數據是小樣本,具有實時性的要求,而且在線應用具有模型偏移的問題。To take charge of such " divergence " to a minimum, this thesis takes the listed companies of different professions as its research samples, seeks the affecting factors of the dividend policies on the companies of all professions ( taking the requirement that the sample value should be no less than 30 in the t test into account, this thesis focuses only on the factors whether to pay cash dividend and the factors affecting the cash dividend paying levers ) and then compares them with the results from the samples of the listed companies, and explores whether these are some distinctions existing and whether it is of significance to research dividend policies according to the profession that each company belongs to
為了將這種「偏離」控制在最小范圍,本文分別以各個行業的上市公司為研究樣本,尋找影響各個行業上市公司股利政策的因素(考慮到t檢驗時樣本量不小於30的要求,本文只研究影響是否發放現金股利的因素和影響現金股利發放水平的因素) ,並與以全體上市公司為研究樣本所得到的結果進行比較,考察是否有區別,區別是否顯著,從而判斷分行業研究股利政策是否有意義。本文將分以下五章。About the confirmation of the position evaluation factors scale, the thesis has put forward using the fuzzy analytic hierarchy process ( fahp ) to evade subjective factor influence and make the position evaluation system more scientific. the thesis makes use of mathematics tools, such as the standard deviation of the sample, variation coefficient, etc. to analyze the deviation that the judge panel give a mark, and carry on effective correct ing
其中在崗位評價因素權重的確定問題上,提出藉助模糊層次分析法( fahp )的系統理論來規避主觀因素的影響,使崗位評價體系的設計更趨科學化;還利用樣本標準差、變異系數等工具來分析評委打分的偏差,進行有效的糾錯處理。Because of idea parameter model, relative efficiencies of sample dmus, which have fuzzy factors of inputs and outputs, can be identified. and projection analysis, clustering analysis and order - sorting of these dmus can also be made. at the same time, we can upbuild the improved schemes for dmus from inefficiency to efficiency and from fuzzy efficiency to efficiency in view of the decision - maker ' s preconception, which can supply decision - makers with useful information and is the foundation for farther controlling and managing
基於該區間dea參數模型,能識別出輸入輸出指標中含不確定因素的樣本決策單元的有效性,並對樣本決策單元進行投影分析、聚類分析及排序分析;也能在系統評價時考慮決策者的個人偏好,給出決策單元從區間dea非有效到區間dea模糊有效或從區間dea非有效到區間dea有效的改進方案,為決策者提供有用的管理信息,為進一步有效管理和控制決策單元提供科學的依據。The method for retaining sampled units in successive sampling survey for changed probability of selection is introduced. for pps sampling design, a model - design unbiased predictor for the total of a variable for the target population is proposed, and the optimum matching ratio for the predictor under the assumption of unchanged population is calculated. for rhc sampling design, the equation that the optimum matching ration satisfies is given
介紹了連續抽樣中概率發生變化時保留樣本的方法。對于有放回的pps抽樣,在假設的超總體模型之下提出了總體變量總值的模型設計無偏預報量,並計算了總體不變時保留樣本的最優匹配比。對于無放回的rhc抽樣,給出了最優匹配比滿足的方程。We make comparative analysis of technical efficiency, pure technical efficiency and scale efficiency while choosing the year1997and 2000 as the model samples. and we also study the degree of the efficiency departure between banks through variance analysis while comparison of the efficiency is made between state - owned commercial banks and the newly founded banks chronologically
橫向上重點選取1997年和2000年為典型樣本點對各銀行的技術效率、純技術效率和規模效率進行了比較分析;縱向上按時間順序比較了國有銀行和新興銀行的效率,並通過方差分析研究了銀行間的效率偏離程度。Is it possible for an estimator to be biased in finite sample but consistent in large sample
一個估計量是否有可能在有限樣本中是有偏的但又具有一致性?But the number of the soil profile that they used was very small
但是他們僅僅是以有限的樣本進行估計,樣本數量偏低。That ' s to say, the observed erraticity will deviate from that of pure statistical fluctuations only if the events of the studied sample are coming from some new kind of physical processes
也就是說,當erraticity偏離純統計起伏時,所研究的樣本中有些事件是來自於某些新的物理過程。分享友人