What Machine Learning Can Do for You

Link: https://www.soa.org/sections/investment/investment-newsletter/2022/february/rr-2022-02-romoff/

Excerpt:

Some ML algorithms (e.g., random forests) work very nicely with missing data. No data cleaning is required when using these algorithms. In addition to not breaking down amid missing data, these algorithms use the fact of “missingness” as a feature to predict with. This compensates for when the missing points are not randomly missing.

Or, rather than dodge the problem, although that might be the best approach, you can impute the missing values and work from there. Here, very simple ML algorithms that look for the nearest data point (K-Nearest Neighbors) and infer its value work well. Simplicity here can be optimal because the modeling in data cleaning should not be mixed with the modeling in forecasting.

There are also remedies for missing data in time series. The challenge of time series data is that relationships exist, not just between variables, but between variables and their preceding states. And, from the point of view of a historical data point, relationships exist with the future states of the variables.

For the sake of predicting missing values, a data set can be augmented by including lagged values and negative-lagged values (i.e., future values). This, now-wider, augmented data set will have correlated predictors. The regularization trick can be used to forecast missing points with the available data. And, a strategy of repeatedly sampling, forecasting, and then averaging the forecasts can be used. Or, a similar turnkey approach is to use principal component analysis (PCA) following a similar strategy where a meta-algorithm will repeatedly impute, project, and refit until the imputed points stop changing. This is easier said than done, but it is doable.

Author(s): David Romoff

Publication Date: February 2022

Publication Site: Risks & Rewards, SOA

Digital-first life insurance policies see increase in demand, Policygenius

Link:https://www.dig-in.com/news/policygenius-life-insurance-policies-no-medical-exams

Excerpt:

The demand for digital-first life insurance products has grown in the last year, likely related to the COVID-19 pandemic. From October to December 2021, about 56% of applications submitted through Policygenius were for no-medical exam policies compared to January to March 2021, which was only 26%.

Author(s): Kaitlyn Mattson

Publication Date: 4 Feb 2022

Publication Site: Digital Insurance

Police lose hacked therapy center criminal reports after spreadsheet error

Link:https://www.thebharatexpressnews.com/police-lose-hacked-therapy-center-criminal-reports-after-spreadsheet-error/

Excerpt:

The hack into the client database of the private Vastamo psychotherapy center was first exposed on October 21, 2020, when the patient data of tens of thousands of people was stolen and used to blackmail both l company and patients.

Investigators asked each victim to file a criminal complaint, and as of February 2021, more than 25,000 such reports had been submitted. The majority of complaints were lodged at the Pasila police station in Helsinki, but others were lodged elsewhere in the country.

….

Instead of a database, criminal reports were saved via Microsoft Excel files. Some of the files turned out to be unreadable when the police attempted to transfer them into the official system. The cause of the problem is unknown.

Detective Inspector Jari Illukka from the Helsinki Police Department told Svenska Yle that a dozen crime reports had disappeared from Excel, but the exact number is not known.

….

Police estimate that the records of more than 30,000 people were stolen during the Vastaamo data breach, and more than 22,000 of those victims have since reported the crime.

However, a little more than three thousand declaration forms had been given to the police at the end of January, that is to say one victim in ten.

Publication Date: 7 Feb 2022

Publication Site: Bharat Express News

Getting Started with Julia for Actuaries

Link:https://www.soa.org/digital-publishing-platform/emerging-topics/getting-started-with-julia/

Graphic:

Excerpt:

Sensitivity testing is very common in actuarial workflows: essentially, it’s understanding the change in one variable in relation to another. In other words, the derivative!

Julia has unique capabilities where almost across the entire language and ecosystem, you can take the derivative of entire functions or scripts. For example, the following is real Julia code to automatically calculate the sensitivity of the ending account value with respect to the inputs:

When executing the code above, Julia isn’t just adding a small amount and calculating the finite difference. Differentiation is applied to entire programs through extensive use of basic derivatives and the chain rule. Automatic differentiation, has uses in optimization, machine learning, sensitivity testing, and risk analysis. You can read more about Julia’s autodiff ecosystem here.

Author(s): Alec Loudenback, FSA, MAAA; Dimitar Vanguelov

Publication Date: October 2021

Publication Site: SOA Digital, Emerging Topics

WordXLe: Wordle in Excel

Link:https://sysmod.wordpress.com/2022/01/25/wordxle-wordle-in-excel/

Excerpt:

If you play Wordle daily, or the French version LeMot, you might want to practice more often. For fun, I created an Excel version that you can download, WordXLe. It has sheets for both English and French versions. The dictionaries are:


English main (for validation) c.12000 words from the SOWPODS dictionary; for play c.1100 words.


Version française: principal c.8000 mots; https://github.com/hbenbel/French-Dictionary/blob/master/dictionary/dictionary.txt

Le jeu 1700 mots. https://www.freelang.com/dictionnaire/dic-francais.php
I removed all accents to simplify the game.


It uses Conditional Formatting for colouring, Data Validation to enforce some letter entry rules, no VBA macros, just formulas. The sheets are protected, but it’s easy to unhide things in Excel if you really want to so I’ll leave that as a challenge. 

Author(s): Patrick O’Beirne

Publication Date: 25 Jan 2022

Publication Site: sysmod

Microsoft Excel: The Program’s Designer Reveals The Secrets Behind The Software That Changed the World 25 Years Ago

Link:https://www.thedailybeast.com/microsoft-excel-the-programs-designer-reveals-the-secrets-behind-the-software-that-changed-the-world-25-years-ago

Excerpt:

In a year when big names from the digital realm profoundly affected the world—Mark Zuckerberg or Julian Assange, take your pick—it’s appropriate to add one more: Douglas Klunder. While largely unnoticed, 2010 marked the 25th anniversary of perhaps the most revolutionary software program ever, Microsoft Excel, and Klunder, now an unassuming attorney and privacy activist for the American Civil Liberties Union in Washington state, gave it to us.

…..

For Doug Klunder, the mission 25 years ago wasn’t so grandiose. As lead developer of Excel, he was handed the job of vaulting Microsoft—then known best for MS-DOS, the operating system in IBM’s PCs—to the forefront in business applications. “We decided it was time to do a new, better spreadsheet,” recalls Klunder, now 50, who joined Microsoft straight out of MIT in 1981 (part of the interview process included lunch with Bill Gates and Steve Ballmer at a Shakey’s pizza parlor).

…..

Klunder and his team came up with “intelligent recalc,” an approach where the program updated only the cells affected by the data change rather than all the formulas in the spreadsheet. Klunder credits Gates with the idea for how to implement the feature—though he says Gates eventually told him he hadn’t implemented what he had in mind at all. Klunder thinks Gates misremembered the discussion, but adds, “Maybe he actually did have a more brilliant idea that now is lost forever.”

Author(s):Thomas E. Weber

Publication Date:14 July 2017 (originally published 2010)

Publication Site: Daily Beast

Standard Chartered fined £46.5m by Bank of England over reporting failures

Link: https://www.theguardian.com/business/2021/dec/20/standard-chartered-fined-bank-of-england-pra

Excerpt:

The Bank of England has fined Standard Chartered £46.5m for repeatedly misreporting its liquidity position and for “failing to be open and cooperative” with the regulator.

The Bank’s Prudential Regulation Authority (PRA) said Standard Chartered had made five errors in reporting an important liquidity metric between March 2018 and May 2019, which meant the watchdog did not have a reliable overview of the bank’s US dollar liquidity position.

…..

One of the errors occurred in November 2018, as a result of a mistake in a spreadsheet entry. A positive amount was included when a zero or negative value was expected, leading to an $7.9bn (£6bn) over-reporting of the bank’s dollar liquidity position.

Author(s): Joanna Partridge

Publication Date: 20 Dec 2021

Publication Site: The Guardian

Emerging Technologies and their Impact on Actuarial Science

Link: https://www.soa.org/globalassets/assets/files/resources/research-report/2021/2021-emerging-technologies-report.pdf

Graphic:

Excerpt:

This research evaluates the current state and future outlook of emerging technologies on the actuarial profession
over a three-year horizon. For the purpose of this report, a technology is considered to be a practical application of
knowledge (as opposed to a specific vendor) and is considered emerging when the use of the particular technology
is not already widespread across the actuarial profession. This report looks to evaluate prospective tools that
actuaries can use across all aspects and domains of work spanning Life and Annuities, Health, P&C, and Pensions in
relation to insurance risk.
We researched and grouped similar technologies together for ease of reading and understanding. As a result, we
identified the six following technology groups:

  1. Machine Learning and Artificial Intelligence
  2. Business Intelligence Tools and Report Generators
  3. Extract-Transform-Load (ETL) / Data Integration and Low-Code Automation Platforms
  4. Collaboration and Connected Data
  5. Data Governance and Sharing
  6. Digital Process Discovery (Process Mining / Task Mining)

Author(s):

Nicole Cervi, Deloitte
Arthur da Silva, FSA, ACIA, Deloitte
Paul Downes, FIA, FCIA, Deloitte
Marwah Khalid, Deloitte
Chenyi Liu, Deloitte
Prakash Rajgopal, Deloitte
Jean-Yves Rioux, FSA, CERA, FCIA, Deloitte
Thomas Smith, Deloitte
Yvonne Zhang, FSA, FCIA, Deloitte

Publication Date: October 2021

Publication Site: Society of Actuaries, SOA Research Institute

Top Excel experts will battle it out in an esports-like competition this weekend

Link:https://www.pcworld.com/article/559001/the-future-of-esports-is-microsoft-excel-and-its-on-espn.html

Graphic:

Excerpt:

Move over, League of Legends. Does anyone even care about Overwatch? No, the real future of esports is spreadsheets and Microsoft Excel. Don’t believe us? Then tune in to ESPN3 or YouTube this weekend to find out.

No, this isn’t a joke. The Financial Modeling World Cup will be held this weekend entirely in Microsoft Excel. And the finals (the quarterfinals, semifinals, and the final match) will all be broadcast live as they happen at 9 AM PT. Everyone’s playing for a total prize of $10,000 — funded by Microsoft, of course.

Author(s):Mark Hachman

Publication Date: 10 Dec 2021

Publication Site: PC World

Emerging Technologies and their Impact on Actuarial Science

Link:https://www.soa.org/resources/research-reports/2021/emerging-technologies-and-their-impact-on-actuarial-science/

Full report: https://www.soa.org/globalassets/assets/files/resources/research-report/2021/2021-emerging-technologies-report.pdf

Graphic:

Excerpt:

Technologies that have reached widespread adoption today:
o Dynamic Collaboration Tools – e.g., Microsoft Teams, Slack, Miro – Most companies are now using this
type of technology. Some are using the different functionalities (e.g., digital whiteboarding, project
management tools, etc.) more fully than others at this time.
• Technologies that are reaching early majority adoption today:
o Business Intelligence Tools (Data Visualization component) – e.g., Tableau, Power BI — Most
respondents have started their journey in using these tools, with many having implemented solutions.
While a few respondents are lagging in its adoption, some companies have scaled applications of this
technology to all actuaries. BI tools will change and accelerate the way actuaries diagnose results,
understand results, and communicate insights to stakeholders.
o ML/AI on structured data – e.g., R, Python – Most respondents have started their journey in using
these techniques, but the level of maturity varies widely. The average maturity is beyond the piloting
phase amongst our respondents. These are used for a wide range of applications in actuarial functions,
including pricing business, modeling demand, performing experience studies, predicting lapses to
support sales and marketing, producing individual claims reserves in P&C, supporting accelerated
underwriting and portfolio scoring on inforce blocks.
o Documentation Generators (Markdown) – e.g., R Markdown, Sphinx – Many respondents have started
using these tools, but maturity level varies widely. The average maturity for those who have started
amongst our respondents is beyond the piloting phase. As the use of R/Python becomes more prolific
amongst actuaries, the ability to simultaneously generate documentation and reports for developed
applications and processes will increase in importance.
o Low-Code ETL and Low-Code Programming — e.g., Alteryx, Azure Data Factory – Amongst respondents
who provided responses, most have started their journey in using these tools, but the level of maturity
varies widely. The average maturity is beyond the piloting phase with our respondents. Low-code ETL
tools will be useful where traditional ETL tools requiring IT support are not sufficient for business
needs (e.g., too difficult to learn quickly for users or reviewers, ad-hoc processes) or where IT is not
able to provision views of data quickly enough.
o Source Control Management – e.g., Git, SVN – A sizeable proportion of the respondents are currently
using these technologies. Amongst these respondents, solutions have already been implemented.
These technologies will become more important in the context of maintaining code quality for
programming-based models and tools such as those developed in R/Python. The value of the
technology will be further enhanced with the adoption of DevOps practices and tools, which blur the
lines between Development and Operations teams to accelerate the deployment of
applications/programs

Author(s):

Nicole Cervi, Deloitte
Arthur da Silva, FSA, ACIA, Deloitte
Paul Downes, FIA, FCIA, Deloitte
Marwah Khalid, Deloitte
Chenyi Liu, Deloitte
Prakash Rajgopal, Deloitte
Jean-Yves Rioux, FSA, CERA, FCIA, Deloitte
Thomas Smith, Deloitte
Yvonne Zhang, FSA, FCIA, Deloitte

Publication Date: SOA

Publication Site: October 2021

How persuasive chatbots might be used in insurance

Graphic:

Excerpt:

Individuals have a different kind of relationship with insurance than what they have with any other product or service. Though being the most effective risk mitigation tool, it still requires a hard push from insurers and regulators to make people purchase. The thought of insurance could evoke every other emotion except joy in an individual. The main reason for this is that insurance is a futuristic promise that assures compensation when a covered risk event happens. This operates exactly opposite to the strong impulse of scarcity and immediacy bias.

As in any other industry, the persuadable events in insurance could be based on reactive or proactive triggers to encourage positive or discourage negative events. Depending on the intelligence ingrained in the back-end systems and the extent customer data is consolidated, the proactive persuasion events could be personalized to a customer and not just limited to generalized promotion of a new product or program. It could be performed for other persuadable events of the same policy for which the chat is in progress or expand to include policy events from other policies of the customer.

An indicative list of the persuadable events in an insurance policy could be categorized as given in Table 2.

Author(s): Srivathsan Karanai Margan

Publication Date: September/October 2021

Publication Site: Contingencies

Excel autocorrect errors still plague genetic research

Link: https://cosmosmagazine.com/science/biology/excel-autocorrect-errors-still-plague-genetic-research/

Graphic:

Excerpt:

Earlier this year we repeated our analysis. This time we expanded it to cover a wider selection of open access journals, anticipating researchers and journals would be taking steps to prevent such errors appearing in their supplementary data files.

We were shocked to find in the period 2014 to 2020 that 3,436 articles, around 31% of our sample, contained gene name errors. It seems the problem has not gone away, and is actually getting worse.

Author(s): Mark Ziemann, Deakin University and Mandhri Abeysooriya, Deakin University

Publication Date: 27 August 2021

Publication Site: Cosmos magazine