The log transformation is special

Department of Medical Statistics, Glaxo Research and Development Ltd., Greenford, Middlesex, U.K.
Statistics in Medicine (Impact Factor: 2.04). 04/1995; 14(8):811-9. DOI: 10.1002/sim.4780140810
Source: PubMed

ABSTRACT The logarithmic (log) transformation is a simple yet controversial step in the analysis of positive continuous data measured on an interval scale. Situations where a log transformation is indicated will be reviewed. This paper contends that the log transformation should not be classed with other transformations as it has particular advantages. Problems with using the data themselves to decide whether or not to transform will be discussed. It is recommended that log transformed analyses should frequently be preferred to untransformed analyses and that careful consideration should be given to use of a log transformation at the protocol design stage.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Even moderate doses of alcohol cause considerable impairment of motor coordination, an effect that substantially involves potentiation of GABAergic activity at δ subunit-containing GABAA receptors (δ-GABAARs). Here, we demonstrate that oxytocin selectively attenuates ethanol-induced motor impairment and ethanol-induced increases in GABAergic activity at δ-GABAARs and that this effect does not involve the oxytocin receptor. Specifically, oxytocin (1 µg i.c.v.) given before ethanol (1.5 g/kg i.p.) attenuated the sedation and ataxia induced by ethanol in the open-field locomotor test, wire-hanging test, and righting-reflex test in male rats. Using two-electrode voltage-clamp electrophysiology in Xenopus oocytes, oxytocin was found to completely block ethanol-enhanced activity at α4β1δ and α4β3δ recombinant GABAARs. Conversely, ethanol had no effect when applied to α4β1 or α4β3 cells, demonstrating the critical presence of the δ subunit in this effect. Oxytocin had no effect on the motor impairment or in vitro effects induced by the δ-selective GABAAR agonist 4,5,6,7-tetrahydroisoxazolo(5,4-c)pyridin-3-ol, which binds at a different site on δ-GABAARs than ethanol. Vasopressin, which is a nonapeptide with substantial structural similarity to oxytocin, did not alter ethanol effects at δ-GABAARs. This pattern of results confirms the specificity of the interaction between oxytocin and ethanol at δ-GABAARs. Finally, our in vitro constructs did not express any oxytocin receptors, meaning that the observed interactions occur directly at δ-GABAARs. The profound and direct interaction observed between oxytocin and ethanol at the behavioral and cellular level may have relevance for the development of novel therapeutics for alcohol intoxication and dependence.
    Proceedings of the National Academy of Sciences 02/2015; 112(10). DOI:10.1073/pnas.1416900112 · 9.81 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This study examined salivary cortisol levels in couples in which one member had unexplained chronic fatigue (CF). The couples completed questionnaires and seven household activities in a laboratory setting and provided salivary cortisol samples prior to and immediately after the activities, as well as again after completing additional questionnaires and debriefing. The couples rated their interactions as similar to those at home, suggesting ecological validity, and patients with CF experienced the activities as involving more exertion than did their partners. The multilevel model results indicated that patients with CF had overall lower cortisol levels and flatter slopes across repeated measurements than did their significant others. Patients' and significant others' cortisol concentrations were significantly associated with each other over time. Furthermore, significant others' cortisol was associated with greater relationship satisfaction and greater observed rates of patients' illness/pain behaviors per minute, but patients' levels of cortisol were not associated with relationship variables. This study is the first to examine cortisol in couples with CF; the results are discussed in terms of implications for future research. (PsycINFO Database Record (c) 2015 APA, all rights reserved).
    Journal of Family Psychology 04/2015; 29(2):296-301. DOI:10.1037/fam0000074 · 1.89 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Recently, various ensemble learning methods with different base classifiers have been proposed for credit scoring problems. However, for various reasons, there has been little research using logistic regression as the base classifier. In this paper, given large unbalanced data, we consider the plausibility of ensemble learning using regularized logistic regression as the base classifier to deal with credit scoring problems. In this research, the data is first balanced and diversified by clustering and bagging algorithms. Then we apply a Lasso-logistic regression learning ensemble to evaluate the credit risks. We show that the proposed algorithm outperforms popular credit scoring models such as decision tree, Lasso-logistic regression and random forests in terms of AUC and F-measure. We also provide two importance measures for the proposed model to identify important variables in the data.
    PLoS ONE 02/2015; 10(2):e0117844. DOI:10.1371/journal.pone.0117844 · 3.53 Impact Factor


Available from