Are you caught up in the machine learning forecasting frenzy? Is it reality or more hype? There's been a lot of hype about using machine learning for forecasting. And rightfully so, given the advancements in data collection, storage, and processing along with technology improvements, such as super computers and more powerful
The U.S. Marshals Service is the federal agency known for bringing wanted fugitives to justice. Often, the Marshals Service gets attention for these arrests, but once the publicity has died down they face a basic challenge --- where to put the individuals in their custody. The agency uses data to
In seinem Buch „Competing on Analytics“ benennt Tom Davenport die Analytik als Grundlage nachhaltiger Wettbewerbsvorteile. Der Grund dafür ist der prädiktive Ansatz. Heutzutage ist es nicht mehr möglich, ein Unternehmen alleine mit Blick in den Rückspiegel zum Erfolg zu führen. Und Analytik erlaubt den dringend erforderlichen Blick in die Zukunft.
Machine learning is taking a significant role in many big data initiatives today. Large retailers and consumer packaged goods (CPG) companies are using machine learning combined with predictive analytics to help them enhance consumer engagement and create more accurate demand forecasts as they expand into new sales channels like the
"Correlation does not imply causation.” Does that bring back memories from your college statistics class? If you cringe when you hear those words, don’t worry. This phrase is still relevant today, but is now more approachable and easier to understand. Here at SAS, we use SAS® Visual Analytics to make
It was John Allen Paulos who said, “Data, data everywhere, but not a thought to think.” That rings true more than ever before. Companies are struggling with the deluge of data coming at them from multiple channels. But traditional data channels are just the beginning. Companies also are facing an
Händler und Handel haben heutzutage Zugang zu einer enormen Menge an Daten – und damit die Grundlage für eine personalisierte Ansprache, die Kunden inzwischen erwarten. Richtig eingesetzt, kann Analytics der Schlüssel für alle möglichen Geschäftsvorteile sein – sei es, dass es darum geht, ein besseres Online-Erlebnis für den Kunden zu
It is said that everything is big in Texas, and that includes big data. During my recent trip to Austin I had the privilege of being a judge in the final round of the Texata Big Data World Championship, a fantastic example of big data competitions. It felt fitting that
Langsam könnte man meinen, das Christkind sei der verbeamtete Chef des statistischen Himmelsamtes. Heute geht es an die Analyse von Zeitreihen zur Vorhersage (Forecasting) der Wunschzettelmassen. Die Kindergesellschaft ist eine Konsum- und Wegwerfgesellschaft. Das merkt das Christkind nicht erst seit 15 Jahren! Da wird gewünscht und ausgepackt was das Zeug
The Rule of Three is a writing principle that suggests that things that come in threes are inherently funnier, more satisfying, or more effective than other numbers of things – Wikipedia. 3 Ps of success, Blind Mice, Little Pigs, Stooges, Musketeers, The Matrix, The Lord of the Rings, rings, pairs
ビジネスに使える「良い予測結果」を得るために 今回は、8月にリリースされたSAS Forecast Serverの新しい機能を紹介しながら、データが理想どおりに画一的にはなっていない実際のビジネスの現場で「良い予測結果」を得るために必要な３つの要素についてご紹介します。 はじめに 需要予測はもともと、天候などの不確実なばらつきをもつ外的要因という制約のもとで、顧客満足度や販売機会を最大化（欠品による損失の最小化）しつつ、売れ残りや在庫保有といったコストを最小限にするための手段のひとつです。生産から販売までのリードタイムが長い商品の販売量の予測や、一定期間先の需要量の取引を行うエネルギーの売買に携わる企業にとっては、正確な需要予測が不可欠です。今回は、予測結果そのものの精度をビジネス上の課題解決に見合う精度にするために、欠かせない要素ついてご紹介します。 欠かせない三つの工夫 SASは長年、SAS/ETSやSAS Forecast Serverなど、時系列予測機能を提供してきましたが、それらツールを使用して実際に成果を出している企業に共通するのは、これらのツールに用意されている時系列予測アルゴリズムを単に使用するのではなく、精度を高めるためのなんらかの"工夫"をしているということです。それをまとめると以下の３つに集約されます。 予測対象の実績データ（以下、時系列データ）のセグメンテーション マルチステージ（他段階の）予測モデリング 予測結果の追跡 予測のためのアルゴリズムは世の中に多数存在しますが、それを単純に適用するだけでは、ビジネス上の意思決定に利用可能な精度を実現するのは実は困難です。従来は、上記３つの工夫をシステム構築の際に考慮し独自に仕組みを作りこむ必要がありました。SASはこれらをベストプラクティス化し、ツールそのものの機能としてリリースしました。新たにSAS Forecast Serverに備わった、それら３つの機能について簡単にご紹介します。 １．時系列データのセグメンテーション どの店舗でいつ何が売れるのかを予測しなければいけない小売業 客室の埋まり方を予測しなければならないホテル業 顧客満足度を落とさずに欠品率をコントロールするために、どの施設レベルで予測すべきかに頭を悩ませる流通業 エネルギーの使用量を予測しなければならないデータセンター事業者やエネルギー供給企業 乗客数や交通量を予測する航空関連企業 コールセンターの需要を予測して従業員の配置を計画する通信会社 など、どのような予測業務においても、最初のステップは、自社の時系列データがどのようになっているかを理解することです。 理想的な時系列データ 従来の予測技術にとって最も完璧な（うれしい）時系列データはこのような形（図１）をしています。量が多く、データ期間が長く、安定していて、同じパターンが繰り返され、欠損値がほとんどなくパターンが予測しやすいという特徴があります。 このような理想的な時系列データが仮に存在したとすれば、自動化された予測エンジンと単一の予測モデリング戦略で簡単に良い予測結果が得られます。 実世界の時系列データ しかし現実世界では、企業が保有する時系列データはもっと多様です（図２）。 洗剤のようにいつでも売れる安定した（Stable）需要 バーベキューセットのように季節性（Seasonal）のある需要 あるいはレベルシフト(Level Shift)があるような需要－例えば、市場や販売チャネルを拡大したタイミングなど－ 新製品の投入や新市場への進出など、データ期間が非常に限定されている（Short History）。 自動車の特定の補修部品のようにスパースなデータ（あるいは間歇需要とも言います）（Intemittent）。 ハロウィングッズのように一年に一回のある週や月しか売れないものもあります（Holiday）。 このようにそれぞれまったく異なる時系列パターンに対して単一の予測モデリング戦略を適用しても良い予測結果は得られません。これらにどのように対処するかが「良い予測結果」を得られるかどうかの分かれ道になります。 時系列データのセグメンテーション この問いに対する解決策は、時系列データのセグメンテーションです。時系列データのセグメンテーションとは、時系列データのパターンに応じて異なるパターンに分類する方法です。これは、需要予測プロセスにおいて最も重要な最初のステップのひとつです。分類した後にそれぞれのパターンに応じた予測モデリング戦略を適用することが「良い予測結果」を得るための秘訣となります。 これにより、 Stableなデータに対しては、ロバストなARIMAモデルを適用し、 季節性を示すデータには季節性モデルを、 Level Shiftタイプにはレベルシフトの要素を説明変数に利用できるARIMAX手法を、 新製品パターンには類似性分析のテクニックを用い、 スパースなデータに対しては間歇需要のための予測モデルを、 Holidayパターンにはカスタマイズした時間間隔モデルを使用する
This guest blog post comes from Dr. David Dickey, one of our original SAS Press authors. Hope you enjoy! In the late 1970s, shortly after SAS was founded, I was approached by Herbert Kirk and John Brocklebank from SAS to put together a course on time series. This was reasonably
Healthcare IT News recently published an article on 18 health technologies poised for big growth, a list culled from a HIMSS database. The database is used to track an extensive list of technology products that have seen growth of 4-10 percent since 2010, but have not yet reached a 70
Healthcare IT News recently published an article on 18 health technologies poised for big growth, a list culled from a HIMSS database. The database is used to track an extensive list of technology products that have seen growth of 4-10 percent since 2010, but have not yet reached a 70
In October I will be at the Analytics 2015 conference in Las Vegas. I’ve never been to Las Vegas before. People tell me that if you are better than average in forecasting where a small ball will end up after it’s been spinning for a while in a dish with
Machine learning is all about automating the development process for analytical models. One way to extend the use of machine learning is to broaden your library of machine learning algorithms. Another way is to scale your machine learning process by reducing the time required to process machine learning algorithms on
If you know me, you know two undeniable things (other than my love for froyo): I consider shopping a sport and I am an Analytics geek. Being an Analytics geek means that I see potential for using data everywhere, and never more than when it’s my data as a customer.
SAS/IIF Grant to Promote Research on Forecasting For the thirteenth year, the International Institute of Forecasters, in collaboration with SAS, is proud to announce financial support for research on how to improve forecasting methods and business forecasting practice. The award for the 2015-2016 year will be (2) $5,000 grants. The
With all the enhancements in demand management over the past decade, companies are still faced with challenges impeding the advancement of demand-driven planning. Many organizations are struggling with how to analyze and make practical use of the mass of data being collected and stored. Others are perplexed as how to
I was recently asked by a customer if they should move the responsibility for creating the statistical baseline forecast. They were considering moving it from their regional country offices to their global headquarters. In addtion, they were considering changing the role of their regional demand planners to only make adjustments to
The date of Easter influences our leisure activities Different from many other public holidays, Easter is a so-called movable holiday. This means that the Easter bunny brings more than just eggs for the statistician - he brings special Easter forecasting challenges. In the year 325 CE the Council for Nicea
Well okay maybe you can't hear us, but at least you can read what we have to say. Although, I'm not ruling out an occasional video or podcast entry. One of the best kept secrets at SAS is the incredible domain expertise that comes to work here everyday, ready to
Bookies have long turned a trade in predicting the fate of our politicians in the general election. According to Ladbrokes, gamblers are set to spend a staggering £100m betting on this year’s result. The outcome of the May 7 vote is anticipated to be the hardest election to predict in
Charlie Chase is considered an expert in sales forecasting, market response modeling, econometrics and supply chain management. Now he's sharing some of his expertise in his Business Knowledge Series (BKS) course, Best Practices in Demand-Driven Forecasting. I had the chance to ask him some questions about his course and the
Once upon a time: The toy industry has invited me to the world‘s largest toy fair, which took place recently in the city of Nuremberg. With close to 3,000 exhibitors the toy fair is bigger than ever before. Success is the theme of the event, and most German retailers cannot complain with consecutive
Why do people steal ATMs? Because that's where the money is!!! While the old "smash-n-grab" remains a favorite modus operandi of would-be ATM thieves, the biggest brains on the planet typically aren't engaged in such endeavors (see Thieves Steal Empty ATM, Chain Breaks Dragging Stolen ATM, An A for Effort). And of
“Let’s assume a normal distribution …” Ugh! That was your first mistake. Why do we make this assumption? It can’t be because we want to be able to mentally compute standard deviations, because we can’t and don’t it that way in practice. No, we assume a normal distribution to simplify
The Forecasting Savant Suppose you received an email from a self-proclaimed forecasting savant, advising you of a big upset in the upcoming mayoral election...and it turns out to be correct. You then get an email picking the underdog in the next championship boxing match...which is right again. Over the course
Many vendors claim they have analytics, and a lot of users have embraced the belief that analytics is the way to go. But what does analytics really mean, especially to business users without statistics backgrounds, and how much do they need to know about analytics to be able to make
Perhaps forecasting is a little of both, crystal ball and competitive edge. It’s a crystal ball of sorts because it helps leaders get answers to questions like, “How many? Or, “How much?” to decide what actions best help the business. And it’s definitely a competitive edge when it results in
