| 释义 |
regression|rɪˈgrɛʃən|
[ad. L. regressiōn-em, n. of action f. regress-, regredī: see regress v.]
†1. Return to a subject. = regress n. 4. Obs.
c1520Barclay Jugurtha (ed. 2) 20 Nowe wyll I make regression and prosecute my first purposed mater insuing mine authour Salust. 1620T. Granger Div. Logike 317 To digression is added also regression, which is a returning back againe to our former speech interrupted by digression.
†2. Recurrence or repetition (of a word or statement). Obs. rare.
1553T. Wilson Rhet. 109 b, That is called regression, when we repeate a worde eftsones, that hath been spoken, and rehersed before. 1597J. King On Jonas (1618) 37 His reason of flying to Tarshish, is againe specified, with a regression in the end of the verse, that he might goe from the presence of the Lord.
3. a. The action of returning to or towards a place or point of departure.
1597A. M. tr. Guillemeau's Fr. Chirurg. 28 b/1 In such accidents wherin is onlye required a regressione of bloode. 1620Venner Via Recta viii. 179 Through the regression of the spirits and heat into the interiour parts. 1682Sir T. Browne Chr. Mor. (1756) 44 Run not into extremities from whence there is no regression. 1864Bowen Logic vii. 225 My going upstairs is my progress towards my object, and my coming down is a regression.
b. Geom. Return of a curve.
1727–38Chambers Cycl. s.v. Point, If the curve turn back again towards the Point whence it first set out, the Point of the Flexure is..called the Point of Regression, or Retrogradation. 1842De Morgan Calculus 434 One sound writer..has attempted to translate the words arête de rebroussement in English by ‘edge of regression’. 1879Thomson & Tait Nat. Phil. I. i. §148 When the number is infinite, and the surface finitely curved, the developable lines will in general be tangents to a curve... This curve is called the edge of regression.
4. a. Return to or into a state or condition; relapse; reversion to a less developed form. Cf. regress 3 a.
1646Sir T. Browne Pseud. Ep. (1650) 29 That essence, which substantially supporteth them, and restrains them from regression into nothing. 1822–34Good's Study Med. (ed. 4) IV. 285 Dr. Home..completed a radical cure in fourteen of them, no relapse occurring notwithstanding the frequency of such regressions. 1882Nature XXVII. 170 The destructive process is identical. It is a regression from the new to the old. 1917Jrnl. Genetics XXVII. 117 What is frequently called reduction in evolution, or, to use a less ambiguous term, regression. 1950Sci. News XV. 136 They pointed out that the regression of tumours caused by Compound E did not generally last indefinitely. The tumours usually recurred. 1965B. E. Freeman tr. Vandel's Biospeleology xxvi. 417 Regression of the eyes is more marked when it is phyletically ancient.
b. Genetics. The tendency for the mean value of a partially inherited quantitative character, among any class of relatives of an individual or a group chosen for their values of that character, to lie between the (mean) value for that individual or group and the mean value in the general population. Esp. as regression to the mean.
1885F. Galton in Nature 24 Sept. 507/1 The experiments showed further that the mean filial regression towards mediocrity was directly proportional to the parental deviation from it. 1889Galton Nat. Inheritance vii. 103, I trust it will become clear..that the law of Regression in Stature refers primarily to Deviations. 1912J. A. Thomson Heredity (ed. 2) ix. 321 The amount of the regression affords a useful measure of the intensity of the inheritance. If the regression is slight, it means that the intensity of the inheritance is high. 1952Srb & Owen Gen. Genetics xxiii. 497 This phenomenon, in which the progeny of selected parents slip back toward the average of the population from which the parents were chosen, has long been known. In this connection the phenomenon was called regression. Today, the term regression has a broader statistical connotation, but it is still applicable in its original sense to the problems we are discussing. 1975A. Smith Human Pedigree iii. 66 At this stage it is necessary to refer to the phenomenon known as the regression to the mean. Where random mating exists, and where there is no evolutionary pressure favouring any characteristic that is controlled by many genes, the offspring will have a tendency to be nearer average for that characteristic than their parents.
c. Statistics. The relationship between the mean value of a random variable and the corresponding values of one or more other variables; coefficient of regression = regression coefficient in sense 8 below.
1897K. Pearson in Phil. Trans. R. Soc. A. CLXXXVII. 259 The coefficient of regression may be defined as the ratio of the mean deviation of the fraternity from the mean off⁓spring to the deviation of the parentage from the mean parent. Ibid., From this special definition of regression in relation to parents and offspring, we may pass to a general conception of regression. Let A and B be two correlated organs (variables or measurable characteristics) in the same or different individuals, and let the sub-group of organs B, corresponding to a sub-group of A with a definite value a, be extracted. Let the first of these sub-groups be termed an array, and the second a type. Then we define the coefficient of regression of the array on the type to be the ratio of the mean-deviation of the array from the mean B-organ to the deviation of the type a from the mean A-organ. 1917Phil. Mag. XXXIV. 205 When the regression of the first variable on the remaining n-1 variables is linear, the multiple correlation coefficient measures the dependence of the first variable on the others. 1925R. A. Fisher Statistical Methods for Res. Workers v. 114 The following qualitative examples are intended to familiarise the student with the concept of regression. 1943M. G. Kendall Adv. Theory Statistics I. xiv. 328 In this chapter we shall mainly be concerned with the case in which regressions are linear or very nearly so. 1952C. G. Lambe Elem. Statistics vii. 56 This straight line which gives an estimate of the average value of y associated with any value of x is called the line of regression of y on x and p/σ 2 / x is called the coefficient of regression of y on x. 1972T. H. & R. J. Wonnacott Introd. Statistics for Business & Economics xiii. 287 Multiple regression is the extension of simple regression, to take account of the effect of more than one independent X variable on the dependent variable Y.
d. Psychol. The process of regressing, or a tendency to regress, in the sense of regress v. 2 b; spec. the tendency of the libido, under the stress of frustration, to return to a simpler and more satisfying stage of development; also, the state of returning mentally to an earlier period, esp. in hypnosis and psychoanalysis.
1910tr. S. Freud in Amer. Jrnl. Psychol. XXI. 214 The flight from the unsatisfying reality into what we call..disease, but which is never without an individual gain in pleasure for the patient, takes place over the path of regression, the return to earlier phases of the sexual life, when satisfaction was not lacking. 1913C. G. Jung On Psychoanalysis in XVIIth Internat. Congr. Med. §xii. 68 [Freud] called this phenomenon of reactivation or secondary exaggeration of infantile reminiscences ‘Regression’. 1920Challenge 21 May 44/3 The libido..in its regression to the collective unconscious, gives rise to the similation of archaic psychical adaptations. 1948Jrnl. Nerv. & Mental Dis. CVII. 443 In regression to infantile levels the subject assumed the sleeping posture of an infant. 1961Economist 11 Mar. 962/1 The poor layman who has laboriously got on to nodding terms with infantile sexuality, regression, Oedipal conflicts, displacement and the rest. 1970T. X. Barber LSD, Marihuana, Yoga & Hypnosis vi. 255 Under regression to infancy, the hypnotized person does not topple from his chair. 1971Jrnl. Gen. Psychol. Apr. 208 In psychology the term ‘regression’ refers to a primitivization of behavior. 1978Gris & Dick New Soviet Psychic Discoveries ix. 105 Let me amplify on regressions. Whatever people think, their previous lives are not individual experiences.
5. Philos. = regress n. 5.
1637R. Humfrey tr. St. Ambrose Pref., The heathen philosopher..holds from privation to habite regression to bee impossible. 1886A. Weir Hist. Basis Mod. Europe (1889) xii. 472 Truths of science are made contingent on a first cause, or are swallowed up in the mysteries of infinite regression.
6. Astron. = regress n. 6.
1823Woodhouse Astron. (ed. 2) II. 660 The annual regression of the Moon's node will be found to be 19°.19′.43{pp}. 1839Moseley Astron. xxxvii. 121 This annual regression of the equinoctial point..is called the Precession of the Equinoxes.
7. Geogr. A retreat or withdrawal of the sea from the land.
1908W. J. Sollas et al. tr. Suess's Face of Earth III. 364 Every new transgression (regression), is so far as the encroaching line of breakers itself has not denuded the land, will encounter an altered relief. 1937Bull. Amer. Assoc. Petroleum Geologists XXI. 1436 Rhythmic transgressions and regressions of the sea continued throughout the period of Jackson sedimentation, evidenced by the interwedging of marine and non-marine sediments, as the strand line moved..back and forth. 1975Sci. Amer. Feb. 90/3 The stratification of sedimentary deposits suggested successive marine transgressions onto the continents and regressions from them. The regressions could be attributed to the subsidence of the ocean basins and the transgressions to the partial filling of the basins with sediment eroded from the continents.
8. attrib. and Comb., as (sense 4 c) regression analysis, regression formula, regression function, regression theory; regression coefficient, a coefficient in the regression equation; esp. the first-order coefficient, which is estimated by the covariance of the two variables divided by the variance of the independent variable; regression curve, a graph of the expected value of the dependent variable plotted against the value of the independent variable(s); regression equation, an equation which gives the expected value of the dependent variable as a function of the value(s) of the independent variable(s); regression line = regression curve above.
1948New Biol. IV. 36 We can then use the technique of regression analysis to determine to what extent we can account for the variation in yield in terms of variation in rainfall. 1976National Observer (U.S.) 19 June 18/3 It may seem absurd to label the probability of murder as ‘P(M)’ and subject it to the technique of regression analysis.
1903Phil. Trans. R. Soc. A. CC. 20 Not only the slope (regression coefficient) of the line, but its position is identical. Ibid. 21 If an organ has been modified only by indirect selection, then its partial regression coefficients on any complex of other organs, however large or small, provided it includes all the directly selected organs, will remain unchanged by the selection. 1925R. A. Fisher Statistical Methods Res. Workers v. 114 The regression coefficients are of interest and scientific importance in many classes of data where the correlation coefficient, if used at all, is an artificial concept of no real utility. 1964R. von Mises Math. Theory Probability & Statistics xi. 576 The correlation coefficient is the geometrical mean of the two regression coefficients.
1905Res. Mem. Drapers' Co. XIV. 21 Yule's method of approaching the problem from the form of the regression curves is..available and capable of very great extension. 1925R. A. Fisher Statistical Methods Res. Workers v. 114 The function which represents the mean height at any age is termed the regression function of height on age; it is represented graphically by a regression curve, or regression line. 1943M. G. Kendall Adv. Theory Statistics I. xiv. 327 The means of arrays will in general lie more or less closely round smooth curves... Such curves are called regression curves and their equations..are called regression equations. 1972G. P. Beaumont Elem. Math. Statistics xii. 152 We begin with a certain minimal property of the regression curves.
1897Proc. R. Soc. LX. 480 The characteristic or regression equations which we have to find. 1943Regression equation [see regression curve above]. 1978Nature 18 May 184/2 The proportion of non-scientific staff [A]..is determined largely by the total staff employed (S) and the number of addresses amongst which it is dispersed (D) according to the regression equation A = 22·23 log S-1·86D-8·77.
1971World Archaeol. III. 115 The regression formula predicts that a compound with five adults will have approximately ten huts.
1925Regression function [see regression curve above].
1904Biometrika IV. 139 The actual degree of resemblance, our brothers being equally variable, is measured by the steepness of this regression line. 1925Regression line [see regression curve above]. 1971World Archaeol. III. 112 Variation away from the regression line is a function of wealth that is not used in this way.
1967Times Rev. Industry Feb. 111/3 The lay reader should not be put off by the complicated language of regression theories, as Professor Kaldor's ideas are fascinating enough, and well enough argued in this lecture to be quite comprehensible to the non-economist. |