Dear Helena,

Confidence interval is actually a matter of Fuzzification where the probability of occurrence of an event could be near to 1 but not exactly 1. Exact 1 i.e. 100% is deterministic, however near to 1 or 100% (i.e. 99% or 95% or 90% etc.) is probabilistic which is non ideal.

100% confidence is a Certainty with a factor of 1 that can only be possible when we take all the observations into account, but in sampling it is not possible to cover the whole population. So in my opinion, Confidence Interval should not be 100% in account.

I also agree with Sir James R Knaub.

Regards.

Helena -  

I am not certain that I understand the question, but in a confidence interval, for there to be a 100% coverage, all possibilities would need to be included.  With the full range as possibilities, that is, only then could you have 100% confidence. Thus, say, if you are willing to have less 'confidence' that your point estimate result would fall in a given interval, you can have a shorter interval.  You become less and less 'confident' about the more and more precise interval. 

 But I think that when talking about rain, a weather report is just talking about the probability that rain will fall in a given geographic location.  That is not statistical confidence, just a somewhat subjective probability based on weather models for the area. If you look at a larger area, say all of Portugal, then the probability of rain is larger that it will rain in at least one of those areas.  That is like looking at reliability for each part of a machine, and then looking at overall reliability of the whole machine.

Cheers - Jim 

Dear Helena,

Confidence interval is actually a matter of Fuzzification where the probability of occurrence of an event could be near to 1 but not exactly 1. Exact 1 i.e. 100% is deterministic, however near to 1 or 100% (i.e. 99% or 95% or 90% etc.) is probabilistic which is non ideal.

100% confidence is a Certainty with a factor of 1 that can only be possible when we take all the observations into account, but in sampling it is not possible to cover the whole population. So in my opinion, Confidence Interval should not be 100% in account.

I also agree with Sir James R Knaub.

Regards.

Dear Helena,

In my opinion the C Lewis kausel answer is perfect "indicator of the precision of a measurement: how stable is the estimate"... IC95% is usually chosen.

Best regards.

Jean.

Statistics as a science require different confidence intervals for different disciplines. As you have mentioned the case study of rain falls probability, let me bring this article to your attention and to the attention of the followers. Why doesn’t the weather forecast have a confidence interval? Dear @Helena, thanks for sharing the question. 

https://wcuk.wordpress.com/2009/06/22/why-doesnt-the-weather-forecast-have-a-confidence-interval/

Because weather forecasts come from weather models that have input of various uncertainties - I suppose there are some 'subjective' input values that increase uncertainty by an unknown amount, but I'm not a meteorologist - it would seem that one could get perhaps an optimistic confidence interval by a bootstrap or similar replication simulation experiment. 

Dear Hussain, James, Lewis, Perriot, Ljubomir et all, 

Thank you  for your answers.

The real-life situations always have similarities with  statistics. For example, in life if we want to help others in a more efficient way we shoud be more precise, give more specific information beeing in that way more usefull...not  being as a 100% confidence interval!

Have a nice day or night...

Helena

PS: Ljubomir it was very interesting  to see the link you add

Dear Helena,

The reason that confidence interval should not be 100% is because calculating confidence intervals in statistics is to determine an interval of estimate in which a parameter of a population might be,  from a given set of sample data. The better work of estimate is to make the interval as small as possible for a very close pinpointing of the population parameter but with higher probability. Taking 100% confidence interval simply means we did not find an estimate interval of the parameter rather the whole of the area of normal distribution curve, for instance in the case of a normal distribution.

Dear Dr. Helena,

100% confidence level indicates the deterministic nature of an event.  In such a case, there is no role for probabilistic studies as the occurrence of that particular event is certain and offers no scope for predictions in terms of probability. I think this is the reason for not having 100% confidence interval.

In statistics, confidence interval describes the amount of uncertainty associated with a sample estimation of a population parameter, and the percentage associated with it is called a confidence level. 

A confidence level helps to account for the other possible sample results that could have been obtained when one is making an estimate of a parameter using the data from only one sample. If one wants to account for 95% of the other possible results, then the confidence level would be 95%.

Confidence levels, in research, are often in the range of 80% to 99%,with the most common confidence level being 95% and the particular choice of a confidence level depends on the field of study or the journal which will publish the results.

 “Why a confidence interval of 100% is not always chosen?” It is because a confidence interval gives a range of reasonable values for some unknown population parameter. Assume that the desired population parameter is the proportion of university lecturers who own a car. What would the range of proportions have to be in order to become 100% confident that it contains the true, unknown proportion? The interval would have to contain all possible proportions and it would have to go all the way from 0  up to 1(which is equivalent to 0% up to 100%)!  But that is not true for a set of practical proportions.

Dear Colleagues,

Good Day,

"Confidence intervals consist of a range of values (interval) that act as good estimates of the unknown population parameter; however, in infrequent cases, none of these values may cover the value of the parameter. The level of confidence of the confidence interval would indicate the probability that the confidence range captures this true population parameter given a distribution of samples. It does not describe any single sample. This value is represented by a percentage, so when we say, "we are 99% confident that the true value of the parameter is in our confidence interval", we express that 99% of the hypothetically observed confidence intervals will hold the true value of the parameter. After any particular sample is taken, the population parameter is either in the interval realized or not; it is not a matter of chance. The desired level of confidence is set by the researcher (not determined by data). If a corresponding hypothesis test is performed, the confidence level is the complement of respective level of significance, i.e. a 95% confidence interval reflects a significance level of 0.05. The confidence interval contains the parameter values that, when tested, should not be rejected with the same sample. Greater levels of variance yield larger confidence intervals, and hence less precise estimates of the parameter. Confidence intervals of difference parameters not containing 0 imply that there is a statistically significant difference between the populations.

In applied practice, confidence intervals are typically stated at the 95% confidence level. However, when presented graphically, confidence intervals can be shown at several confidence levels, for example 90%, 95% and 99%.

Certain factors may affect the confidence interval size including size of sample, level of confidence, and population variability. A larger sample size normally will lead to a better estimate of the population parameter."

Please, see the following articles:

-  Field, Andy (2013). Discovering statistics using SPSS. London: SAGE.

 -  Zar, J.H. (1984) Biostatistical Analysis. Prentice Hall International, New Jersey. pp 43–45

http://en.wikipedia.org/wiki/Confidence_interval

https://www.health.ny.gov/diseases/chronic/confint.htm

http://sphweb.bumc.bu.edu/otlt/MPH-Modules/BS/BS704_Confidence_Intervals/BS704_Confidence_Intervals_print.html

Good question dear Helena.

At first, confidence interval means predictability of occurrence of an event or process. Again, Predictability involves  permutation/combination with uncertainty in every case.  If there is no uncertainty and if the process occurs via only one route then there is no meaning for confidence interval. Because in those case there is no permutation/combination and   you are able to control the whole process. If you are able  to control  any process by yourself then you can say 100% confidence.   

Dear Prof. Doko

Simple and Excellent  answer to the point.

Dear all,

Thank you vey much...in conclusion...we never should say 100% to anything because it is the same of "saying nothing"...has lots of uncertainty...little usefulness..

Helena

In my humble opinion, 100% confidence intervals would not be confidence intervals at all. 100% confidence means we are completely sure that our results match the general population. However, we are not able to collect data from EVERYBODY. Also intervals imply some "space" and 100% certainty leaves no space for any doubt. 

I think it is not only logical but a bit philosophical issue. Hope it helps a bit. 

The possible reason is that a confidence interval to a level of confidence of 100% would be the set of real numbers R or the range of the random values, it would be an obvious result.

Discover Confidence Intervals Research! It brings research, question, answers and notable researchers in this topic.Similar categories are also treated.

https://www.researchgate.net/topic/Confidence-Intervals