Daily Streamflow Time Series Modeling by Using a Periodic Autoregressive Model (ARMA) Based on Fuzzy Clustering

Round 1

Reviewer 1 Report

The article is interesting, the combination of models is a major contribution. From the reading I consider that the following aspects should be expanded:

1) It is necessary to make an extension of the input data, including a statistical analysis (variances, means etc) of the data, this is a help to understand the cluster.

2) A more extensive analysis of the clusters should be made, including an analysis of the locus of the clusters and their values.

3) The analysis of quadratic forecasting should be expanded to include comparisons with other models.

Author Response

Dear Reviewer,

Greetings,

I appreciate your kind revision of the paper. Please kindly check the attachment.

Best Regards,

Mahshid Khazaeiathar

Author Response File: Author Response.pdf

Reviewer 2 Report

see attached file 

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

Greetings,

I appreciate your kind revision of the paper. Please kindly check the attachment.

Best Regards,

Mahshid Khazaeiathar

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The new version includes all the requested aspects

Reviewer 2 Report

The authors have incorporated my comments. However, I have two points:

1- In the title, they add "ARMA" but after "autoregressive" which is not correct. it should be "autoregressive moving average".

2- In Table 4 p and q are model orders (integer) and not coefficients. Please correct or explain want are these values in the table. 

After these corrections the paper can be published.