I have asked you two questions:

How are the graphs you present derived from your data?

Which are the similarities you claim to be there?

It is right, you are measuring temperature. Nevertheless, your graphs are showing frequency.

You present several graphs and claim they show some similarity pattern. I study your graphs and fond no patterns. Presumably, you are able to explain how you detect the claimed pattern.

The answers to these questions do not require knowledge about homeopathy to understand.

..... I apologize for the long time in replying, but I have little time for internet forums at the moment.

Best regards, Hans

__________________
<i>You have a right to your own opinion, but not to your own facts.</i>

Dear Hans I have posted a link to excel file containing the original data. you can have a look at them Identification of Homeopathic Medicines by Measuring Physiological Variability in Human Body Temp

Thanks,

Dr.Devendra Kumar

OK, I got them thanks. I'll put them through a statistical analysis later when I have the time, but just looking at them, I notice that each column varies only within about half a degree.

Since this vairation is within the uncertainty of a normal skin temperature probe, I think you may be looking at system noise.

Hans

__________________
<i>You have a right to your own opinion, but not to your own facts.</i>

Dear hans,

Happy to hear that you got the original data of temparature variability. I hope you can apply various statistical methods apart from AR Spectrum, as you are expert in the feild.

Thanks,

Dr.Devendra Kumar

Dear Devendra Kumar,

I have now tried to analyze your data in various ways. My observations are the following:

1: Preparation and placebo are different, but with low significance, and only if samples are taken as cohorte. This is, however, technically not valid; instead, cohortes should be either sensors or subjects. In this case differences are not statistically significant.

2: The differences between the various preparations and sensors are no smaller than the difference between preparation and placebo. This suggests that differences are within system variability.

3: All measurement series show a normal distribution, with excellent significance (all <0.01). This suggests that most variances are random.

4: Some of the measurement series show a slow change over time. This suggests a normal variation in local skin temperature, and there is no particular correlation between various input parameters.

5: The sampling rate is two seconds. This correlates with the 0.5Hz peaks in the frequency analysis, and suggests that those peaks are an artefact of the measuring interval.

6: The variances within each measurement series is within +-2%. This is within the expectable accuracy of a cutaneous temperature measurement and suggests that most of the variances are system noise.

7: Inter-series variances are within +-3%. This is well within the expectable variance between measurements taken with different sensors, in different positions, and on different individuals.

Conclusions:

- Variances cannot be distinguished from system noise.
- The frequency analysis peaks are an artefact stemming from the sample frequency (the sampling frequency will always show up as peaks in a frequency spectrum analysis).

Things that can be deduced from your experiment: Skin temperatures vary slightly over time, between individuals, sensors, and measuring positions. These deductions are hardly news.

Sorry, that's about it. If you feel I have overlooked something, feel free to point it out.

Best regards, Hans

__________________
<i>You have a right to your own opinion, but not to your own facts.</i>

For further research, I suggest that you test your claim about being able to distinguish between preparation and placebo in a blinded setting:

Have someone code 10 vials of preparation and placebo in a random pattern. Save the key to the coding in a sealed envelope. Perform the recording and see if you can identify the vials correctly. If you get more than 6 right, you may be on to something.

Br, Hans

__________________
<i>You have a right to your own opinion, but not to your own facts.</i>

Mr.Hans,

Thanks for your spended time and for most valuable suggestions.
I will follow the steps you have suggested. Will meet you soon.

As I am working as research asst in govt reserch institute. I am not getting sufficient time to perform my work on large samples.

I will come back soon.

Once again thank you very much for your valuable suggestions

Dr.Devendra Kumar.

You are very wellcome! I appreciate your attempt at objectivity.

The great difficulty in designing tests for homeopathic remedies is the lack of predictivity.

Usually, in science, we have some idea or theory about how something should work, and we can design a test to prove or disprove that theory.

However, nobody knows how homeopathic remedies function (if they function, of course), so any tests to identifiy them will totally arbitrary. And such tests are very difficult to evaluate because if you don't know what to expect it is easy to mistake spurious results for real results.

Best regards and good luck, Hans

__________________
<i>You have a right to your own opinion, but not to your own facts.</i>