Holt-Winters predicts over a 30days period as opposed to months

Barclaeys
Barclaeys New Altair Community Member
edited November 2024 in Community Q&A
hi, I have just built a Holt-Winters model with Apply forecast on monthly aggregated data. My month is of Type "Date" and is shown as Sep 1, 2015 / Oct 1, 2015 / Nov 1, 2015...

When I look at the prediction, the result set does not predict for Jun 1, 2020 / Jul 1, 2020 / Aug 1, 2020 but takes Jun 1, 2020 / Jul 30, 2020 / Aug 29 2020 / Sep 28 2020 so it seams to keep a fixed 30 days rather than doing the forecast for the 1st of the month. 

Is this normal and is there any way I can change this?
Thank you!
Tagged:

Best Answer

  • MartinLiebig
    MartinLiebig
    Altair Employee
    edited August 2020 Answer ✓
    it somewhat is. Those methods assume, that your data is equidistant. Sadly months (and years) are not equidistant. So what the algorithm takes is i think the average of the steps, since we prefer to give you a solution over an error.
    I think @tftemme and team is working on finding a nice solution for those "calendar types". For now there is not a real solution, other than using a generic ID, like I do in my Auto-Forecasting project: https://community.rapidminer.com/discussion/comment/66543#Comment_66543

    Alternativly one can think about a postprocessing to adjust the date to the next 1st of the Month.

    Best,
    Martin

Answers

  • MartinLiebig
    MartinLiebig
    Altair Employee
    edited August 2020 Answer ✓
    it somewhat is. Those methods assume, that your data is equidistant. Sadly months (and years) are not equidistant. So what the algorithm takes is i think the average of the steps, since we prefer to give you a solution over an error.
    I think @tftemme and team is working on finding a nice solution for those "calendar types". For now there is not a real solution, other than using a generic ID, like I do in my Auto-Forecasting project: https://community.rapidminer.com/discussion/comment/66543#Comment_66543

    Alternativly one can think about a postprocessing to adjust the date to the next 1st of the Month.

    Best,
    Martin

  • Barclaeys
    Barclaeys New Altair Community Member
    Martin, thanks for the feedback. this answers my question. Would you have an example of such postprocessing option?
  • MartinLiebig
    MartinLiebig
    Altair Employee
    sure! Attached is an example doing it. This works until you forecast more than 15 months, which is i think kind of rare.
    Best,
    Martin

    <?xml version="1.0" encoding="UTF-8"?><process version="9.7.002">
      <context>
        <input/>
        <output/>
        <macros/>
      </context>
      <operator activated="true" class="process" compatibility="9.7.002" expanded="true" name="Process">
        <parameter key="logverbosity" value="init"/>
        <parameter key="random_seed" value="2001"/>
        <parameter key="send_mail" value="never"/>
        <parameter key="notification_email" value=""/>
        <parameter key="process_duration_for_mail" value="30"/>
        <parameter key="encoding" value="SYSTEM"/>
        <process expanded="true">
          <operator activated="true" class="utility:create_exampleset" compatibility="9.7.002" expanded="true" height="68" name="Create ExampleSet" width="90" x="45" y="136">
            <parameter key="generator_type" value="attribute functions"/>
            <parameter key="number_of_examples" value="100"/>
            <parameter key="use_stepsize" value="false"/>
            <list key="function_descriptions">
              <parameter key="date" value="date_add(date_parse_custom(&quot;01/01/2010&quot;,&quot;dd/MM/yyyy&quot;),id,DATE_UNIT_MONTH)"/>
              <parameter key="value" value="rand()"/>
            </list>
            <parameter key="add_id_attribute" value="true"/>
            <list key="numeric_series_configuration"/>
            <list key="date_series_configuration"/>
            <list key="date_series_configuration (interval)"/>
            <parameter key="date_format" value="yyyy-MM-dd HH:mm:ss"/>
            <parameter key="time_zone" value="SYSTEM"/>
            <parameter key="column_separator" value=","/>
            <parameter key="parse_all_as_nominal" value="false"/>
            <parameter key="decimal_point_character" value="."/>
            <parameter key="trim_attribute_names" value="true"/>
          </operator>
          <operator activated="true" class="time_series:arima_trainer" compatibility="9.7.000" expanded="true" height="103" name="ARIMA" width="90" x="246" y="136">
            <parameter key="time_series_attribute" value="value"/>
            <parameter key="has_indices" value="true"/>
            <parameter key="indices_attribute" value="date"/>
            <parameter key="p:_order_of_the_autoregressive_model" value="1"/>
            <parameter key="d:_degree_of_differencing" value="0"/>
            <parameter key="q:_order_of_the_moving-average_model" value="1"/>
            <parameter key="estimate_constant" value="true"/>
            <parameter key="main_criterion" value="aic"/>
          </operator>
          <operator activated="true" class="time_series:apply_forecast" compatibility="9.7.000" expanded="true" height="82" name="Apply Forecast" width="90" x="380" y="136">
            <parameter key="forecast_horizon" value="5"/>
            <parameter key="add_original_time_series" value="true"/>
            <parameter key="add_combined_time_series" value="true"/>
          </operator>
          <operator activated="true" class="generate_attributes" compatibility="9.7.002" expanded="true" height="82" name="Generate Attributes" width="90" x="514" y="136">
            <list key="function_descriptions">
              <parameter key="delta_this_month" value="date_diff(&#10;date,date_set(date,1,DATE_UNIT_DAY)&#10;)/1000/60/60/24"/>
              <parameter key="delta_next_month" value="date_diff(&#10;date,&#10;&#10;date_add(&#10;&#9;date_set(&#10;&#9;&#9;date,1,DATE_UNIT_DAY&#10;&#9;),&#10;&#9;1,&#10;&#9;DATE_UNIT_MONTH)&#10;&#9;&#10;)&#9;/1000/60/60/24"/>
              <parameter key="adjusted_date" value="if(&#10;&#9;abs(delta_this_month)&lt;abs(delta_next_month),&#10;&#9;date_set(date,1,DATE_UNIT_DAY),&#10;&#9;date_add(date_set(date,1,DATE_UNIT_DAY),1,DATE_UNIT_MONTH)&#10;&#9;)"/>
            </list>
            <parameter key="keep_all" value="true"/>
          </operator>
          <connect from_op="Create ExampleSet" from_port="output" to_op="ARIMA" to_port="example set"/>
          <connect from_op="ARIMA" from_port="forecast model" to_op="Apply Forecast" to_port="forecast model"/>
          <connect from_op="Apply Forecast" from_port="example set" to_op="Generate Attributes" to_port="example set input"/>
          <connect from_op="Generate Attributes" from_port="example set output" to_port="result 1"/>
          <portSpacing port="source_input 1" spacing="0"/>
          <portSpacing port="sink_result 1" spacing="0"/>
          <portSpacing port="sink_result 2" spacing="0"/>
        </process>
      </operator>
    </process>


  • Barclaeys
    Barclaeys New Altair Community Member
    Just beautiful. And I only copied the XML into the RapidMinder GUI with some tweaks to the expression. Really great.

Categories