LinkedIn

Creating online technology courses for the LinkedIn Learning data science library. Developed and recorded courses on Microsoft Power BI, AWS QuickSight, Excel, R, and Python.

LinkedIn Learning Instructor

Helen Data Design

Interested in how to make data analytics and visualization a large scalable solution to understanding more about your business? I do Microsoft Power BI and AWS freelance consulting and training sessions on all parts of the technology, including queries, ETL frameworks, DAX calculations and modeling, and visualizations.

Founder and Principal

Cornell University

Facilitating courses for predictive analytics, R, Python, and Excel optimization models for data science certifications, including engaging with students, grading projects, and designing additional projects to improve student experiences

eCornell Course Facilitator

Rice University

Technology lead for Python for Business Analytics (MGMT 665) course for MBA students at the Jones Graduate School of Business

Lecturer in Management

Founderz

Developed and presented content in the Power BI-focused chapter in their new course: Online Master in AI and Innovation, created in collaboration with Microsoft.

Professor

CODE Magazine

Authored articles with content focusing on Power BI, Tableau, data visualization best practices, and Excel Power Query

Contributing Author

UMBC Training Centers

Teach courses to government agencies on Python, R, Excel, and machine learning

Data Science Instructor

Accelare

Created Power BI dashboards utilizing design principles on the front end, and DAX languages and multiple data sources on the back end to build out data visualizations to be leveraged by different business departments and levels of users. Provided Accelare consulting staff with training on methods and approaches to leveraging Excel’s functionality to provide more effective and efficient client solutions. Project manager for predictive analytics Medicaid project. Designed and developed Microsoft Power BI metrics designed to be incorporated with Accelare’s WorkFit software that strategy to execution to numerous Fortune 500 companies. Provided the client analytical support through calculation and analysis the modeling of their employment data. Combined this data with allocated cost totals to transform the unstructured view to a useful visualizations, dashboards, and report, assisting the client in identifying significant cost reductions of over $10 million. Real-time logistics analysis with Splunk (SQL structured database) to reduce revenue losses for a large city transportation system by over $3 million a year by improving efficiency for fixing open ticket gates. Analysis and design of HR organizational charts to streamline personnel structure to a lower cost base by cutting around five percent for each department.

Business Architect

Advanced rollout of the OneClaim global implementation in 23 countries, including translation of system information into five languages, for the largest Pegasystems project in company history designed to save AIG $1 billion a year in total operating costs for processing claims. Conceptualized and configured the AIG claims platform, strengthening platform performance to address operational requirements from use cases to the creation and customization of PegaRULES Process Commander (PRPC) requisites. Conferred with stakeholders worldwide about technical and delivery components of implementation projects. Culled requirements for major projects from end users such as the North American Private Client Group and insurance lines in France.

Business Analyst

Part of Actuarial Student program. Calculated and analyzed deferred annuity GAAP and statutory reserve liabilities for financial reports and statements, encompassing an in-depth range of quarterly and annually prepared exhibits. Documented the account value of over $2 billion in deferred annuities as a valuation actuary role. Analyzed financial statements and drafted reports using SAP accounting software, running financial queries and utilizing SAP reports to deepen analysis of the actuarial projections for the deferred annuity block of business. Evaluated and forecasted the deferred annuity cost (DAC) model as part of an annual review designed to align the projected mortality rates, interest rate credited, interest earned, and the expenses in-line with the observed financial experience of the model over the previous year’s performance. Yearly DAC unlocking process reflected in the reported GAAP revenue statements, and these numbers include an increase of $200 million in reported earnings for the entire deferred annuity block of business in 2013. Modeled annuity product interest rates in accordance with market projected interest rates and guaranteed rates. Produced and validated life insurance and annuity illustrations on multiple software programs leveraged by agents to illustrate life insurance/annuity products to a diverse customer base including high net worth clients. Verified product information, specifications, and assumptions accuracy, such as interest rates and fund values. Crafted testing strategies to verify project insurance information accuracy, originating programs to confirm the use of the correct values throughout key processes from pricing to final insurance software calculations.

Actuarial Assistant

Follow me to see my latest published content on data science, analytics, and visualization! Aficionado of Power BI, AWS, Tableau, and Excel (including Power Query), and currently diving deep into the world of Python, R, and AI (including machine learning).  Designer and developer of business intelligence dashboards and visualizations supporting many business functions and industries.

One of the reasons that DAX is a game changer in Power BI is because it enables us to do so many helpful time intelligence calculations like running totals and period-over-period comparisons.

Excel doesn't offer these functions for directly calculating these metrics because it doesn't work in the same way that DAX does. While Excel does offer options for working with arrays and tables, it primarily references data by each cell within the worksheets of a workbook. To reference a range of cells, we need to expand our formulas to an array or create a table within our model to reference within a formula.

DAX, on the other hand, works automatically with columns (like arrays) and tables. When we create a DAX measure formula, for example, we reference the tables and the columns within it that are already part of the data model. If we want to reference a single data point or a value within a summarized table visual, we need to add another step to our DAX formula to reference this value using the MAX function for example.

While DAX often makes some types of calculations much harder to set up (especially ad hoc ones), it enables us to create powerful time intelligence calculations from that eponymous function category that easily references the tables and columns within the existing data model. While I first encountered DAX in PowerPivot in Excel years ago, it became a game changer for me when I learned how to harness it within Power BI.

Here's how date/time functions and time intelligence functions compare between the Excel and DAX languages as Venn diagrams. We can see that for the date and time functions, Excel and DAX share many of the same functions, but several functions work in one or the other, but not both. (As a note for the Venn diagram below, DATEDIFF and the older DATEDIF function are similar, but not the same.)

I shared diagrams like this in the past, but date and time calculations are so important that they're worth continually sharing different analyses and perspectives on the topic. I'll share more on this topic in the future too, so stay tuned! #Excel #DAX #TimeSeries

Helen Wall

We can run Python scripts in Power Query to run ETL frameworks and create visuals in Power BI, but have you tried doing it with DAX? Check out how to do with the PY function in DAX formulas to create tables, columns, and measures in your Power BI model! #PowerBI #DAX #Python

Edited April 2nd: Happy April Fool's Day! The functions PY and XL are available in Excel, more to come on this topic!

Looking to reduce the number of DAX measures in your Power BI model? Calculation groups in Power BI work like a wrapper around existing DAX measures by helping us scale out models over common filters like time intelligence calculations for running and rolling totals.

Here's how we can take 28 DAX measures in a Power BI model and scale it to 4 DAX measures and 7 calculation items in a new calculation group. We can build out this time intelligence calculation group in the Model view of a Power BI model ourselves.

This is also where the SELECTEDMEASURE DAX function becomes a game changer in Power BI! #PowerBI #DAX #DataModeling

The Power BI update for March came out last week! Christopher Wagner, MBA, MVP and I put together our updates for this month along with monthly updates for the last two and a half years combined into one consolidated document.

Chris's top pick for this month: mobile layout auto-create

My top picks for this month: Copilot for DAX (AI with DAX!), data label alignment, matrix data bars

We share these updates every month when the Power BI update comes out - stay tuned for more of these updates to come in the future! #PowerBI #MonthlyUpdate #AlwaysBeLearning

Power BI Updates - March 2024

What makes a data visualization effective? If we look at Tableau's five most influential visuals of all time, they all center around some of the most pressing and important topics over time.

Many of these data visualizations focus on who, what, when, and where questions surrounding the data. They're most effective if they answer why the data is important and why a wider audience should pay attention to what the visuals are communicating.

Here's an overview of these visuals on a high level:

1. Chart of Biography (Joseph Priestly): Who were historical biographical figures and when did these people live?

2. War Mortality (Florence Nightingale): When sanitary measures were implemented in hospitals during the Crimea War, what happened to the UK soldier deaths and why are these measures important?

3. March on Moscow (Charles Minard): When Napoleon invaded Moscow in 1812, what happened to the French army, and why was it a terrible idea?

4. Gapminder (Hans Rosling):  What happened to global trends in life expectancy and income when analyzing these trends over the last 200 years by country and aggregating region?

5. London Cholera Map (John Snow): Where was cholera occurring in London in 1854, what was the common factor for these cases, and why was contaminated water the cause of this epidemic?

What makes these visuals impactful, though, is not just what their designers included, but also what they chose to take out or not include to focus on the most important points that they wanted to communicate! #DataVisualization #Tableau #WorldHistory

Ever thought of SQL joins as zippers?

A lot of SQL learning content uses Venn diagrams as a way to illustrate how different query join types compare. I found that when I first learned SQL many years ago, however, the concept of these overlapping circles didn't make a lot of sense to me at first.

Instead, I preferred to think of them as zippers, where the teeth that matched up between the sides of the zippers were a metaphor for the values on the sides of the analogous key dimension field that appears in the tables on both sides of the join. Later, once writing code for joins in SQL became second nature, the Venn diagrams then made a lot more sense to me.

I'm putting materials together for a SQL project that I'm working on right now, and I came across one of my favorite diagrams on this topic in the process. I updated it to a darker background to share again. I hope others find this helpful as well! #SQL #DataScience #AlwaysBeLearning

When building Power BI models, we can take different approaches to storing and managing data. For example, we can use Power Query to bring in the data using an ETL framework. In this scenario, the data might come from a larger repository source, like a database. In some cases, though, we can reduce the overall model size by using DAX measures instead.

For example, I'm working with solar calculations in a few projects right now to calculate the sunrise and sunset times by location. In the Power Query example for the data model on the left, we can see how bringing in each calculated sunrise and sunset time for 200 locations over a single year gives us over 70,000 data records, which along with the rest of the Power BI model, takes up around 2 MB of space.

Conversely, if we do the same calculations using DAX measures instead, we see on the right that we can reference the smaller date and location tables, which in turn make the model a fraction of the size of the other one (61 KB instead). While the visuals seemed to take a split second longer to load in this model, the lag wasn't really noticeable either.

While the solar calculations for the sunrise and sunset times are scalable, we can't scale every model in this way. This is just one example of how DAX measures can drastically decrease the overall size of the model and enable us to scale it out in ways that we can't otherwise do very easily by importing the data directly through Power Query. #PowerBI #DAX #PowerQuery

Have you ever wanted to work in an industry specifically because of an office building? Count me as part of that cohort.

When I was little, I saw a photo of the modern, gleaming Lloyd's of London building in an encyclopedia. Opened in 1986 and designed by the high-tech style architect Richard Rogers, it stands out even amongst the City of London skyline for its unique metal exterior that pushes the utilities infrastructure to the outside of the building.

Founded in 1688 in a coffee shop on Tower Street in London by Edward Lloyd, it started as a place for merchants to obtain marine insurance for shipping and cargo. It's not actually an insurance company either, but rather a global hub marketplace for insurance and reinsurance where syndicates come together to pool and spread risk. Like much of the insurance world in general, the building also conceals more than it reveals.

Several years ago, I was in London for a work trip, and someone in the office I was working at kindly offered to give me a tour of the building (it's not open to the public otherwise). The building more recently graces the London skyline, but the history and artifacts inside it go back centuries like a time machine.

The Lutine Bell in the middle of the trading floor historically simultaneously notified all the brokers and underwriters at the same time about a ship's status. It rang once to indicate the bad news of a ship loss, and twice for the good news that an overdue ship just returned safely.

It's also surreal to go into the Adam Room on the 11th floor, which contains the reassembled opulent dining room that once belonged in the large Bowood estate house (built in the 1700s, demolished in 1956). It contrasts quite a lot with the modern expansive open-space office with lofty ceilings for trading on the floors below that.

I haven't worked in insurance in a long time, but this is one of my favorite memories of working in the industry, aside from people I worked with. Thank you Dan Brodnitz for reminding me of my favorite London work trip moment when you visited London last week! #Lloyds #Insurance #Architecture

As with any language, there are functions that we use far more than other ones. Within the DAX language, one function that I don't use very often is SELECTEDMEASURENAME. But here's an example of how to use it to avoid having to tediously repeatedly write out the filter condition in the second parameter of the DAX CALCULATE function!

Instead of setting the filter for the Fruit field of the Data table to Apples, Bananas, Oranges, or Pears respectively, I set the name of each measure equal to the fruit name and referenced the measure name as the filtered value.

Admittedly, this is a limited use case example, and it can run into issues when we use this approach in larger models because it doesn't tell us that we're trying to calculate the total sales for each respective fruit. Rather it just gives us the fruit name for each respective measure name. But for a recent example of my own where I didn't want to cumbersomely write out the individual measure conditions for each measure iteration, this approach saved me a lot of time! #DAX #PowerBI #DataAnalytics

Looking to maximize the capabilities of the new card visual in Power BI? Check out how to format the labels for all the measures together or separately to make it user-friendly and easy-to-read.

In a bit of a mini-series within the Power BI Weekly series that comes out every week, I use this new card visual (still in the preview features mode!) to create a calendar heat map for the selected month of data. In this example, I use weather data for temperatures, but the data can be virtually any type of time series data.

Check out the latest video in this serial course in the LinkedIn Learning library! Also, check back every week on Wednesdays for five-minute video content on topics and tricks like this that I wish that I knew earlier myself! #PowerBI #DataVisualization #TimeSeriesModels

Easter this year is on March 31st, but next year it will be on April 20th. Why is there such a discrepancy between these dates, even just a year apart? The answer lies in "computus," or the ancient algorithm calculations that determine this date. Easter occurs on the first Sunday after the first full moon after the spring equinox. Below you can see how to calculate this date using Python code!

However, there are discrepancies even within these calculated date parameters. For example, the ecclesiastical full moon is different from the full moon that astronomers calculate today. Also, it uses March 21st as the spring equinox, even though the equinox typically occurs a day or two before then. "Computus" also uses the Gregorian calendar to calculate dates, while Orthodox Easter uses the Julian calendar, which typically falls on a later date (in 2025, though, they will both line up on April 20th).

I find solar and lunar calculations like this fascinating, and recently I decided to learn more about them! It's also fascinating how Abrahamic religions all use slightly different calendars based on the moon's cycles to determine Passover and Ramadan for example as well. #Python #Easter #TimeIntelligence

I'm always looking for new and interesting ways to visualize data, and this visual is one example of something that recently caught my eye! With time series data modeling, for example, we often use line charts to visually represent the data points over time.

I was watching a Vox video on extreme weather forecasting when I came across this graphic. While it doesn't provide exacting data points like a line chart does, it emphasizes another equally important part of forecasting: the ambiguity that goes with it farther into the future. I like the way it shows the higher level of uncertainty of future model outcomes with increased pixelation of the Earth's surface and the weather models projected onto it. #TimeSeriesModels #Forecasting #DataVisualization

The new card visual in Power BI is quite extensive and versatile. So much so, in fact, that I have several videos in the Power BI Weekly series dedicated to leveraging different aspects of it.

This week, I'm covering how to add customized labels to each card in the visual. This is helpful in scenarios like a calendar because the day within a month changes as we select different month-year combinations.

New to LinkedIn Learning? Check it out for free with the link below! #PowerBI #DAX #LinkedInLearningInstructor

The new card visual in Power BI offers a lot of possibilities for creative data visualizations. One example of this is a calendar visual, which enables us to select a year and month to filter the data to display within the calendar.

It takes a bit of work to set up this visual, which I separated across several videos in my Power BI Weekly serial course. Once the visual comes together, it loads and updates quite quickly to reflect the month the end user selects in the month-year slicer.

This week, I cover how to set up the DAX measures for the values on the cards in any selected calendar month. In the weeks to come, I'll share how to update the individual card labels, their borders, value formatting, and the color gradients. Stay tuned for more to come on this topic!

New to LinkedIn Learning? Check out this video (plus the rest of this course and thousands more) in the library with the link below! #PowerBI #DataVisualization #DAX #LinkedInLearningInstructor

The new Python in Excel function enables us to integrate Python code directly into an Excel workbook. To do so, we need to use functions within Excel formulas to bridge the languages.

Along with converting between Python and Excel outputs, there are also Excel shortcuts with keyboard key combinations specifically for working in the Python in Excel (PYXL for short) functionalities. Here's a cheat sheet of those specific keyboard shortcuts. More to come on this topic, stay tuned! #Python #Excel #AlwaysBeLearning

Nothing quite emphasizes how fast time passes like viewing dates as serial numbers as we see in Excel when displaying the date values since January 1st, 1900.

I was thinking about this recently when I was doing calculations that involved converting dates into Julian Day and Julian Century values. Using Excel as a hands-on modeling tool for figuring this out placed the serial numbers for dates right in front of me along the way. It's hard to believe today (March 26th, 2024) has a stored date value of 45,377. Every interval of three years contains over one thousand days within it. On July 6th, 2009, the date serial number was 40,000. It's amazing how fast time passes and analyzing time series daily really emphasizes this! #Excel #TimesSeries #DataAnalytics

Is it just me or has the preview features menu in Power BI become so long that it requires either using the scroll bar to navigate or vertically extending the options menu? It's exciting to see so many features in preview mode, but it's also easy to miss some of them at first glance!

I have all the preview features options turned on except on-object interaction (which I personally find makes Power BI workflows more cumbersome) and less-elevated user support (security and sensitivity labels are important!).

I'm curious - which features did others turn on from the April 2024 Power BI Desktop update as a comparison? #PowerBI #MonthlyUpdates #AlwaysBeLearning

Next week on the 3rd April at 18:00 GMT+1, the London Excel Meetup group welcomes Helen Wall to present on Python in Excel.

This is an online event and you can participate via Zoom or YouTube LIVE.

Event link to sign up (It's FREE)
https://lnkd.in/ezicKTTe 

This session will focus on using Python in Excel to extend the capabilities of Excel rather than replacing it. There's a lot of overlap between Excel and Python but deciding what to use can seem daunting.

This presentation will show how to use Excel as much as possible and then how a bit of Python code can enhance models rather than completely reinvent them.

#pythoninexcel #londonexcelmeetup #dataanalysis

meetup.com • 1 min read

PYXL: To use Python in Excel or not? | Helen Wall, Wed, Apr 3, 2024, 6:00 PM | Meetup

1:41 P.M. yesterday afternoon (April 8th, 2024) in Dallas, Texas as the solar eclipse encased the region in darkness for several minutes.

It's difficult to describe what an incredible experience it is to see an eclipse in the path of totality. I certainly didn't realize it myself until yesterday.

For the experience, I also wore an appropriate shirt and certainly wore the approved actual sun-safe eyewear (with photographic evidence). But I also had to be diligent about swapping these special glasses with my regular sunglasses (the former were hard to see with when walking around outside)!

Learning more about solar calculations over the last few months certainly helped further pique my interest in this topic, but there's so much out there that I still want to learn about these phenomenons. #SolarEclipse2024 #SolarCalculations #InRealLife


You know the feeling. You said yes to something, and now it’s time to do it. But instead of feeling excited, you feel dread. This actually has a name: it’s called the “Yes-Damn Effect”. But there's a trick.

The Yes-Damn Effect is usually caused by dynamic inconsistency: the “you” that needs to do the thing is not the same as the “you” that agreed to do the thing. 

You probably know this well: it’s what happens when you go to the supermarket when you’re hungry and buy a bunch of junk food, and then when you’re home and not hungry, you regret it.

As every software developer knows, we tend to overestimate how much time our future selves will have, and underestimate how long the things we currently have on our plates will take, and how many random things will pop up between now and then. 

So even if we were able to estimate correctly the benefit of the thing we agreed to do, we tend to underestimate the cost. 

Damn, indeed.

It can be even worse: sometimes we overestimate the benefits, too. But let’s not even dwell on that, because there’s a cool technique that I want to tell you about. It’s called the “No-Yay Effect.”

As you become more experienced, you learn that you need to say “no” sometimes. You can’t do everything, and you can’t do everything well. But saying “no” is hard. You immediately see the disappointment of the person you’re saying “no” to, but you may not see the benefit to you.

So here’s a cool trick.

When you say “no” to something, put an empty reminder on your calendar for the time when you would have done the thing. A very simple way is to put a 30-minute meeting with yourself on the day that the request was due. I usually mark that time as “free”, so it doesn’t actually block my calendar, but I make sure to turn on the reminder.

This will start to give you some positive reinforcement for saying “no” as you’ll start to see the benefit of the time you’ve saved. It’s not uncommon that one such reminder will pop up in my calendar when I’m really busy and then I think “Yay, I’m glad I said no to that thing.” 

Yay, indeed.

Over time, this positive reinforcement may make it easier for you to say “no” to things. And you may be happier for it.

The original version of this post was in my blog:
https://lnkd.in/ehiEtKG4

Taplio

Helen Wall's Linkedin Analytics

Helen Wall

What is Helen talking about?

Helen Wall's Best Posts (last 30 days)

Want to drive more opportunities from LinkedIn?

Taplio

.css-1jgreyo{background-image:linear-gradient(284deg, rgba(39,43,72,1) 0%, rgba(33,47,183,1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Helen Wall's Linkedin Analytics

Helen Wall

What is Helen talking about?

Helen Wall's Best Posts .css-qqfgvy{font-size:var(--chakra-fontSizes-md);}(last 30 days)

Want to drive more opportunities from LinkedIn?

Helen Wall's Linkedin Analytics

Helen Wall's Best Posts (last 30 days)