Distribution Fitting Tool - finding of best fitting distribution type

2 ビュー (過去 30 日間)
gayan wijendrasiri
gayan wijendrasiri 2020 年 10 月 19 日
コメント済み: Jeff Miller 2020 年 10 月 21 日
I'm just trying to analyze soma data using distribution fiting tool of Matalab. I want to find the mean by fiting some curve types like Log, exponential, Burr, etc. I manually calculated the mean of my data set and compared with the mean produced by the Matlab relative to various curve types. Sometimes though the curve is fitted well, the means are different. As per manual calculation mean is 0.588175. The mean values proposed for various types of curve fiting options are mentioned below.
  1. Exponential - 0.588175
  2. Bur - 1.06174
  3. Log-Logistic - 0.686126
How can I find the best fitting curve for my dataset? Further I extracted the [x,f(x)] data for the fitted curve from the matalab and poltt the curve in excel. Though the curve is ok the mean calculated from these values are different to the mean produced by the matlab. As an example Bur distribution indicates a value of 1.06174 for mean, but the extracted data gives a mean arround 0.5.
Your valuable comments are highly appreciated since I am still new to this tool.

採用された回答

Jeff Miller
Jeff Miller 2020 年 10 月 19 日
It depends on exactly what you mean by "best fitting curve". In general, there is certainly no guarantee that the best-fitting parameters for each distribution (e.g., by maximum likelihood) will give you means that match the means in your data. You will see even bigger discrepancies if you check the standard deviations of the distributions against the standard deviation in your data (e.g., for the exponential). Your only option is to pick your optimization criterion (e.g., maximum likelihood); once you do that, you just have to live with the discrepancies in predicted/observed means, standard deviations, etc. If you really want to match those and don't care about likelihood, you can use the method of moments for your optimization criterion, but the estimates obtained with that method don't generally have nice statistical properties.
Hope that helps.
  2 件のコメント
gayan wijendrasiri
gayan wijendrasiri 2020 年 10 月 21 日
Thnak you very much.
Jeff Miller
Jeff Miller 2020 年 10 月 21 日
You are welcome. If this answer is what you needed, please accept it so the question will be closed.

サインインしてコメントする。

その他の回答 (0 件)

カテゴリ

Help Center および File ExchangeProbability Distributions についてさらに検索

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by