English translation: fit and likelihood statistics
Login or register (free and only takes a few minutes) to participate in this question.
You will also have access to many other tools and opportunities designed for those who have language-related jobs (or are passionate about them). Participation is free and the site has a strict confidentiality policy.
Spanish to English translations [PRO] Science - Mathematics & Statistics / health care
Spanish term or phrase:estadísticos de ajuste
I need some urgent help to translate this phrase and, in fact, to understand the meaning of the latter part of this sentence:
"Se calculó el tamaño de muestra necesario para analizar el modelo multivariable de asociación entre variables con un nivel de confianza del 95%, y cumplir el teorema de Moivre de frecuencia estimada mayor que cinco y para que los estadísticos de ajuste y de verosimilitud pudieran explicarse como una ji-cuadrado."
My translation so far is: "We calculated the sample size required to analyse the multivariate association model between variables with a confidence interval of 95% significance, and to fulfill the de Moivre theorem of estimated frequency greater than five and to express the adjustment and verisimilitude statistics as a chi-squared.
To be quite honest I'm out of my depth with this paragraph and don't know if what I've written makes any sense at all. Help! Any statisticians out there?
Explanation: A statistic (countable noun, as opposed to the science of statistics) is a number that tells us something about a sample, such as its average value, mean, regression coefficients, etc.
Also, "likelihood" is a more suitable translation of "verosimilitud" than "verisimilitude" when used in this context; see link below for details or check Google for the relative frequencies of both when coupled with "statistic".
Also, I would translate the "ji-cuadrado" part as "...as a chi-square distribution". I doubt he is talking about the test, but about the fit statistics being the parameters of a distribution.
Thanks very much to both of you for your help and comments. Yes, I think 'fit and likelihood' statistics and 'a chi-square distribution' are probably my best options. I'm going to wait one more day to see if anyone wants to add any further points before grading and closing.
The distribution and the test are slightly different things. There are several tests using the chi-square distribution, which are either suited to specific purposes or corrected, more sophisticated versions of older tests. See frex. the relationship between the Pearson's and the Yates' tests (Wikipedia is, again, surprisingly good here).
And yes, there is nothing wrong with going beyond the original question when relevant!
Urtzi has made that point that there may be a difference between a "Chi-squared test" and a "Chi-squared distribution". Obviously, we're getting beyond the translation of "estadísticas de ajuste" here, and getting into the translation of the entire phrase (although that is probably a good thing!)
I'm good at maths, but my expertise doesn't reach to this level. Doesn't a chi-squared test check whether a set of data has a chi-squared distribution? See my 2nd ref:
"A chi-square test ... is any statistical hypothesis test in which the sampling distribution of the test statistic is a chi-square distribution when the null hypothesis is true, or any in which this is asymptotically true".
Perhaps we should consider that the complete phrase is "estadísticos de ajuste y de verosimilitud" and that the intention is that these can "explicarse como una ji-cuadrado". Note that "una ji-cuadrado" is feminine, so it may refer to a chi-squared distribution (distribución) or a chi-squared test (prueba).
Hava a look at the following:
That's what I thought, too. But Googling "estadísticos de ajuste" (with the quotes) gets over 4000 hits, while "estadísticas de ajuste" only gets 847. The RAE is not really clear on this, but it seems like "la estadística" is the name of this branch of mathematics, while "los estadísticos" refers to "los datos". You live and learn!
One confusing aspect of this phrase is that I thought that estadísticos as a noun could only refer to 'statisticians' ie the person? Isn't 'a statistic' (ie. a particular figure) or statistics (the field) generally referred to as 'la estadística'?
I suspect that "modelo multivariable de asociación entre variables" is a bit redundant. It seems to be taken for granted that in a multivariate model, the associations are between the variables, so it's just "multivariate association model" (Google this and you don't find any refs that add "between variables", etc)
Automatic update in 00:
Answers
11 mins confidence: peer agreement (net): -1
Appropriate statistics
Explanation: Appropriate statistics
We calculated the sample size required to analyse the multivariate association model between variables with a confidence interval of 95% significance, and to fulfill the de Moivre theorem of estimated frequency greater than five and to express the adjustment and verisimilitude statistics defined as chi-square .
Explanation: Please note that "estadísticos" refers to a noun that is not "statistics".
It probably refers to "coeficientes o factores estadísticos".
On a separate note, I think "como una ji-cuadrado" refers to a "chi-square function".
Hope you'll find this to be useful.
Giovanni Rengifo Colombia Local time: 03:19 Works in field Native speaker of: Spanish PRO pts in category: 12
Explanation: A statistic (countable noun, as opposed to the science of statistics) is a number that tells us something about a sample, such as its average value, mean, regression coefficients, etc.
Also, "likelihood" is a more suitable translation of "verosimilitud" than "verisimilitude" when used in this context; see link below for details or check Google for the relative frequencies of both when coupled with "statistic".
Also, I would translate the "ji-cuadrado" part as "...as a chi-square distribution". I doubt he is talking about the test, but about the fit statistics being the parameters of a distribution.