Insights

Outliers in Retained Surgical Devices

Outliers pose problems for many different types of analysis. While in many analyses they can either be removed or even ignored without much impact on the final result. However, in some small counts, they are problematic. Researchers at 5E examined publically available data regarding retained surgical devices to identify whether instances are representative of a systemic problem.

In a recent paper published and presented at the Northeast Decision Sciences Institute Conference in Philadelphia, PA in April,  the team from 5 Element Analytics scrutinized the RSD metric to identify potential systemic issues at hospitals.

Using a similar approach from a previous paper in which a method was used to identify problems with the ERA metric in baseball, this approach discusses how outliers can be identified which would have an impact on hospital performance.

Visit our website or contact us for a copy of the paper, “Small Count Data and Outlier Analysis: An Exploratory Study of Patient Safety”

If you would like more information contact

Kiyra Artisse
Chief Operating Officer
5E Analytics
Phone: 516 945 0923
Email: press@5eanalytics.com

3 Major Concerns for Revenue Forecasting

Revenue Forecasting presents a major challenge for companies, especially in finance and marketing. Erroneous forecasts can be disastrous for a companies bottom line, especially when money is spent in anticipation of higher revenues. While the challenges are numerous, there are some actions a department can take to ensure their forecasts are accurate and continue to be accurate.

1. Bad Data

This won’t come as a surprise to many. Every company has bad data, but does this mean you can’t get an accurate forecast, no. First, the data should be analyzed for the causes of errors. Once the causes are known, the errors can either be removed or accounted for, in a model, greatly enhancing the accuracy.

Not all your data, therefore, is bad. The trick is to identify the data that is good, which will act as a core baseline. This baseline can then be used to identify variations and fluctuations to determine the cause. Factors, such as spikes in class center volume, vacations by salespeople, or temperature all can impact results.

Our data scientists, continually find ways to mitigate the issues of bad data through statistical imputation and simulations. These methods can greatly enhance an understanding of the effect of the bad data, as well as provide methods to remediate the issues.

2. Changing Environment

“Change is the only constant in life”, was a famous quote by Heraclitus. Many revenue forecasting models preach “teach it and leave it”, as though it can continue to provide accurate forecasting without ever modifying it. However, this isn’t really the case. Economic environments change, organizations change strategies, and customers change preferences. In some cases, models can pick up the changes, but more complex models require retraining of the model. While many companies say this can occur automatically, it cannot.

Take for example a scenario in which a variable was never included in the original model. If this variable becomes important at some point in the future, the model could never compute its impact, yielding an incorrect result, such as would occur with a new product category or removal of an entire product category.

We think of this as driving a car, always making adjustments. While generally a smooth ride, there are times in which scenarios appear, where traditional training doesn’t work and we need to make a split-second decision. Algorithms cant generally do this, without massive computing power, such as what is needed for self-automated driving.

Change is the only constant in life – Heraclitus

Revenue forecast models need regular and periodic review for accuracy and training.

3. Bad Interpretation

Ultimately, the models produce a number for the forecast, models that are too general have increased errors overall, while specific models have fewer errors for training sets, and increased errors for data it hasn’t seen. Analysts’ interpretations are critical when this occurs since the end-user of the forecasting model isn’t fully aware of what events are causing fluctuations.

One way forecasters combat this is to create an ensemble model, which aggregates a number of models together. The challenge with this approach is that many there may not be an insight into which part of the ensemble is more weighted, thus the ensemble simply tries to smooth out these fluctuations, causing a loss of, most likely valuable information.

Creating multiple models allows for a more granular interpretation of the forecast as well as greater insight into the error. This examination of error is critical to understanding the events that are impacting the revenue. Simultaneously, it creates a better-disciplined process by which analysts, forecasters, and end-users are diligent in understanding as much of the variation as possible.

Creating forecasting models in a vacuum should be a serious concern for the organization. This black-box approach, without truly understanding the effect of multiple data points and methods can lead to a forecast accurate in some instances, but which can lead to failures down the road. We believe these black-box approaches create more problems, and that a solid repeatable reviewable process is the most reliable approach and yields more opportunities.

Exploring The Flaws in Earned Run Average

Professional baseball teams have long used metrics  to measure player performance. The Earned Run Average (ERA) has been at the cornerstone of baseball metrics, since its inception in 1912.  Traditionalists have abhorred the use of new metrics conceived by Sabrematricians, however, to many they have proven quite valuable.

In a recent paper to be published and presented at the Northeast Decision Sciences Institute Conference in Providence, RI in March,  the team from 5 Element Analytics scrutinized the ERA statistic to help formulate a better approach to identifying pitching performances that would be candidates for exclusion, otherwise known in the statistical world as outliers.

There is a great deal of insight in this research that exposes the impact of these hidden outliers, which can really create a true measure of a pitchers performance-

Tyler Levine, Pitcher for The Long Island Ducks Professional Baseball Club

The new approach discusses how outliers can be identified using statistical methods by comparing the change in variance of a set of observations by the removal  of a single observation. In this way, a single bad game, or even an unusually good game are identified thus allowing for a true measure of performance. The data was initially tested on the top 10 pitchers in the National League, and identified fewer outliers than other methods, but found a better representation of a pitcher’s “True” ERA.

Since earned runs follow a count process, identifying outliers have benefits in a number of different areas.  The approach could have significant implications for use in other areas such as healthcare, and transportation.

We would like to thank Tyler Levine and Long Island Baseball, in Bellmore, NY for their support during this research.

Click Here for a copy of the paper , “Outier Identification of Count Data Using Variance Difference”, being presented in Providence, RI, in April 2018 .

Should you like more information contact

Kiyra Artisse
Chief Operating Officer
5 Element Analytics
Phone: 516 945 0923
Email: press@5eanalytics.com

Analytics – A Sun Tzu Perspective

Many of Sun Tzu’s writings have been adapted in a number of industries such as investing, real estate, and sales. However, the application of Sun Tzu to analytics is especially interesting and insightful because it plays such as significant role in successful analytics projects.

Sun Tzu said,

“There are five dangerous faults which may affect a general:

(1) Recklessness, which leads to destruction;

(2) cowardice, which leads to capture;

(3) a hasty temper, which can be provoked by insults;

(4) a delicacy of honor which is sensitive to shame;

(5) over-solicitude for his men, which exposes him to worry and trouble. “

How could one relate anything from analytics to these precepts? Very easily, if you are a student of strategy, and of Sun Tzu.

If an data analyst hurries their work or provides information in manner that is not complete, it leads to the the destruction of the work and possibly the destruction of companies or departments. An analysis that hastily attempts to identify a market segment, marketing campaign or competitive price comparison, can lead to disastrous results. This is not to say that speed isn’t critical, on the contrary as we will see in another article, speed is important, but reckless speed, kills.

Second, an analyst, must have the courage to state what must be said. They cannot hide the information or fear retribution from commanders and leaders. It is their responsibility to give their leadership critical, accurate and timely information whether its good or bad.

Third, analysts must leave the ego at the door. They cannot be tempted to lash out that their solution is the only reasonable one. While confidence and conviction are important, analysts must listen to all inputs in order to ensure that their solutions are not reckless. If there is a rush to judgement or just a defense of ones own position, the battle will be lost.

Fourth, if one is afraid to stand behind their decisions, the risk increases of hesitation and an analysis that is at best incomplete. Analysts must be willing to stand by their work, and be willing to be humble about their results, and not worry about preserving an undeserved honor. Honor and praise will come from decisiveness and completeness. However, a good analyst will shun overt and excessive praise because the work is never done.

Finally, our team is very important and data scientists are very difficult to come by; however, if leaders of analytic teams are worried about the sensitivities of their team or try and make things too comfortable, the analytics team  will never be sharp for difficult projects. Training and goals are necessary for the analytics team to excel in what they do, comfort and complacency are the real enemy.

Sun Tzu’s effect on business spans many disciplines. Analytics teams are no exception. We can learn much from Sun Tzu and in our next articles, we will explore more of how to apply Sun Tzu to your data Science team.

The key to our success….

In our company, we face the normal everyday pressures of businesses. We always work hard for our clients and are continually seeking ways to improve. We don’t let ourselves forget that we are humans as well. We get excited and happy when our ideas become reality and we get frustrated and stressed when we just can seem to find the answers. So how do we overcome periodic downturns of creativity? Continue reading The key to our success….