literally thank you so much its 5 am have spent 3 hours trying to decipher my professor just for this one part and you fixed my issues in less than 10 minutes
Thank you sooo much! After an entire semester in quarantine i felt like I was going to 100% fail Forecasting. But after watching a couple of your videos I finally understand the fundamentals!
Hey can you please tell me which course are you taking, where you study things like forecasting? is it data analytics? I ll be joining university soon and wanted to know.
@@learn_with_rungmangHi, i'm studying Value Chain Management which includes everything from Marketing and Supply Chain to Forecasting, Inventory Management, Planning and many others.
Thank you for great video ! Do you think this model still work perfectly with 12 months data and some month with 0 value. Then is there any way to apply for multi sku products... in the same sheet?
How can I build forecast.ets Excel function essentially by hand so I have more control over inputs? This is for seasonality as well. I ran forecast.ets.stat so I have the smoothing parameters. Do you have videos on the equations so I can reproduce the results of the exponential triple smoothing by hand?
thanks leslie.. quick question.. I followed the steps however final solver step did not work... alpha, beta, gamma and RMSE remained unchanged even after i used the solver to optimize the constants.... any guidance?
Bill, To download the data used by Leslie in this example: 1) visit this page at Redfin: www.redfin.com/news/data-center/ 2) scroll down to "Home Prices, Sales & Inventory" 3) select "All" for States 4) select the "Download" Tab 5) select the "Download" Icon at the bottom of the Table (right side of gray Tableau bar) 6) choose "Crosstab" from the list 7) choose "Excel" from the next list and click "Download" 8) download complete Good luck!
@@JimKernix I see the discrepancy. You can reduce the Home Sales number by filtering by Property Type to "Single Family Residential". However, this doesn't reduce the number to the example value though. My guess is Redfin has expanded their sales database and the downloaded table will still work, but the results will vary from the example. *I have counties around me that still are not tracked by Redfin and not part of their current sales database. FYI...I have used @Leslie Major layout and formulas on a private sales database and it performs the same as in the example. Speaking of that, you can filter the Redfin sales data to your specific area and use her layout for your specific market. Good luck!
Hey Leslie, I have an assignment cool, I wanted to know when we are using this method, do we need to find the initial seasonal factors, or we use the first year current values ?
When we complete the forecast shouldn't the month be off by 1 month due to the plus one. I.e., for your example when you used February should you take the seasonal factor of march?
Hey thanks a lot for the explanation, I just have a question, what if you´re forecasting sales for 2021 and want to make some adjustment due to the pandemic economic shut down in 2020, do you have any advice regarding this situation?
I guess Im asking randomly but does anybody know of a way to log back into an Instagram account?? I stupidly lost my password. I would love any tips you can offer me.
@joyrao4537 that's such a good question. The initial value can be treated different ways. Some people even use solver to solve for the optimal initial values. Do you have a data set with this initial value of 0 to share? I could have a look
Thank you for tutorial, my question, when i have 10 data jun until maret, how i calculate level? Because find level must have y/average(y1,..y12). Please give me answer thank you
for this there are various formula you can use but let me help in answering your question first...the time series variables or observations , for example stock are the y1,y2,y3...yn , for this particular example, sold are the y1,y2,y3...yn..e.g the first observation is the y1 etc ....for levels you also use [(Y3 - Y1)/2] or [(Y2 - Y1) + (Y3 - Y2)/2] ... so to answer your question, let the first observation take the value y1, then use any formula. If I got your question right
Yes, you would have to use 52 periods instead of using 12 periods, since 52 weeks in a year vs 12 months in a year.. And instead of months match week with equivalent week of previous year. Split seasonality into weeks. It will become a lot more tedious though.
Yes if the trend is negative, the data is trending down. Sometimes the trend for a particular timestep doesn't reflect what the actual data is doing. That can happen particularly if there is lots of random fluctuations in the data
Hello Leslie. I love your channel. I am interested to talk to you. I would like to talk to you about integrating excel skills in industrial engineering studies. How can i reach you? thanks
So, if I understand it correctly, up until period = 63 (March) we have the Actuals. And the forecast column tries to replicate (backtest) the Actuals. But I miss the real Forecasts into the future. In my excel I have tried to drag the formula down to forecast 36 months in the future, but it didn't work. How should I procede from here?
![[LIVE] : ONE ลุมพินี 88 | คู่เอก "ป้อมเพชร vs อัสลามจอน"](http://i.ytimg.com/vi/1ZqTVVbMgbo/mqdefault.jpg)
literally thank you so much its 5 am have spent 3 hours trying to decipher my professor just for this one part and you fixed my issues in less than 10 minutes
So much more understandable for me, than my professors explanation away from Excell.
Thank you sooo much! After an entire semester in quarantine i felt like I was going to 100% fail Forecasting. But after watching a couple of your videos I finally understand the fundamentals!
Hey can you please tell me which course are you taking, where you study things like forecasting? is it data analytics? I ll be joining university soon and wanted to know.
@@learn_with_rungmangHi, i'm studying Value Chain Management which includes everything from Marketing and Supply Chain to Forecasting, Inventory Management, Planning and many others.
@@iliankostadinov8209 Thankyou so much! Really appreciate it! :)
Love the way you step thru this and break out the formula. Thank you.
I'm so thrilled to have found this video. Thank you!
You are the best teacher! It's so clear! Thank you!
Wow! could you please share a downloadable excel file link for that one. that would help a lot
Well explained. Breaking it down step by step helps. Thx
thank you so much Leslie. You cooked, I love you
What if you had to forecast for more than one season? Which seasonal factors would you use?
Dear Leslie , shouldn't the forecast for February consider the level and trend values of January ? The formula says so.
Can data be weekly instead of Month in quarter? the Model is still be effective?
Thankyou for this!...can we get this excel sheet? it would immensely help.
Thank you for great video ! Do you think this model still work perfectly with 12 months data and some month with 0 value. Then is there any way to apply for multi sku products... in the same sheet?
This is great, thank you. Is the spreadsheet available?
this is really helpful, thank you! can we have this excel file?
If a company only has past data for 1 year, how should it go about forecasting for future months mitigating the seasonal effects?
Thankyou for helping my exam
How can I build forecast.ets Excel function essentially by hand so I have more control over inputs? This is for seasonality as well. I ran forecast.ets.stat so I have the smoothing parameters. Do you have videos on the equations so I can reproduce the results of the exponential triple smoothing by hand?
I need to forecast 48 months but in this example we only need 9 months to forecast. I don't know how to take seasonalities can you please help?
This was very helpful, thank you!!
Thank you! You explained it so well !
Aww
Sorry. Please answer Mrs, It's aditive or multiplicatice?
thanks leslie.. quick question.. I followed the steps however final solver step did not work... alpha, beta, gamma and RMSE remained unchanged even after i used the solver to optimize the constants.... any guidance?
Is there a way to get your data file?...so I can practice and follow along.?
Bill,
To download the data used by Leslie in this example:
1) visit this page at Redfin: www.redfin.com/news/data-center/
2) scroll down to "Home Prices, Sales & Inventory"
3) select "All" for States
4) select the "Download" Tab
5) select the "Download" Icon at the bottom of the Table (right side of gray Tableau bar)
6) choose "Crosstab" from the list
7) choose "Excel" from the next list and click "Download"
8) download complete
Good luck!
@@metropropertiesgroup4363 Home Prices, Sales & Inventory is not on that page
Found it - the new link is www.redfin.com/news/data-center/ then follow all the steps she mentions
However, the numbers don't match. Homes sold in her spreadsheet for January 2015 is 131646 but the file has 310785.
@@JimKernix I see the discrepancy. You can reduce the Home Sales number by filtering by Property Type to "Single Family Residential". However, this doesn't reduce the number to the example value though.
My guess is Redfin has expanded their sales database and the downloaded table will still work, but the results will vary from the example. *I have counties around me that still are not tracked by Redfin and not part of their current sales database.
FYI...I have used @Leslie Major layout and formulas on a private sales database and it performs the same as in the example. Speaking of that, you can filter the Redfin sales data to your specific area and use her layout for your specific market.
Good luck!
Thank you for the great tutorial
Hey Leslie, I have an assignment cool, I wanted to know when we are using this method, do we need to find the initial seasonal factors, or we use the first year current values ?
When we complete the forecast shouldn't the month be off by 1 month due to the plus one. I.e., for your example when you used February should you take the seasonal factor of march?
can this holt winter method be used in forecasting weekly data?
Hey thanks a lot for the explanation, I just have a question, what if you´re forecasting sales for 2021 and want to make some adjustment due to the pandemic economic shut down in 2020, do you have any advice regarding this situation?
I guess Im asking randomly but does anybody know of a way to log back into an Instagram account??
I stupidly lost my password. I would love any tips you can offer me.
hi thank you so much, this really helps me! where can I find the source of the formula to find the initial value? it's for citation purpose!
great explanation, thank u so much!
Great explanation Thank You..!!
Thank you for your help!
what to do if one of the initial month's sales are 0, as later when I drag the formula I get division by 0 error
@joyrao4537 that's such a good question. The initial value can be treated different ways. Some people even use solver to solve for the optimal initial values. Do you have a data set with this initial value of 0 to share? I could have a look
Thank you for tutorial, my question, when i have 10 data jun until maret, how i calculate level? Because find level must have y/average(y1,..y12). Please give me answer thank you
for this there are various formula you can use but let me help in answering your question first...the time series variables or observations , for example stock are the y1,y2,y3...yn , for this particular example, sold are the y1,y2,y3...yn..e.g the first observation is the y1 etc ....for levels you also use [(Y3 - Y1)/2] or [(Y2 - Y1) + (Y3 - Y2)/2] ... so to answer your question, let the first observation take the value y1, then use any formula.
If I got your question right
Great video, thanks!
This the Holt winter’s additive or multiplicative method?
great work!!! thank you
Thank you!
You're welcome!
what if i wanna model another year? I use "k" as 13,14...24?? I think that overestimates the demand
hey sorry for bothering you after one year but I got the same question, do you mind sharing answer if you got one?
do you have this excel file?
Can we utilise Holt Winters method for weekly forecasting? Anyone? and if yes what formulas would one use?
Yes, you would have to use 52 periods instead of using 12 periods, since 52 weeks in a year vs 12 months in a year.. And instead of months match week with equivalent week of previous year. Split seasonality into weeks. It will become a lot more tedious though.
best video
sorry, I don't understand why the trend forecasting has some negative number, is ít acceptable?
Yes if the trend is negative, the data is trending down. Sometimes the trend for a particular timestep doesn't reflect what the actual data is doing. That can happen particularly if there is lots of random fluctuations in the data
@@lesliemajor6676 oh, so does it go the same with the level and season?
@@lolbich9420 yes although the seasonal shouldn't be negative if the data is well suited for holt winters
@@lesliemajor6676 thanks, now i have understood, thank you so much for answering my question
@@lolbich9420 You're very welcome!
Hello Leslie. I love your channel. I am interested to talk to you. I would like to talk to you about integrating excel skills in industrial engineering studies. How can i reach you? thanks
Good one, just checking have you done holts winter method for daily data using excel. Like I need to calculate my next month day wise, thanks
Hi! Is this the additive or multiplicative Holt Winter's Model?
I think it's multiplicative one.
I think so
its pseudo additive
So, if I understand it correctly, up until period = 63 (March) we have the Actuals. And the forecast column tries to replicate (backtest) the Actuals. But I miss the real Forecasts into the future. In my excel I have tried to drag the formula down to forecast 36 months in the future, but it didn't work. How should I procede from here?
Awesome!! :)
please share your data
Will do! I'll post the file tomorrow:)
@@lesliemajor6676 where did you post the excel file please?