FICO

  Want to protect yourself online - and protect your family - from data compromise and identity theft? FICO's Doug Clare, vice president for cybersecurity solutions, offers some advice in this interview with NBC King 5 News in Seattle. He was interviewed in conjunction with his talk at the US Chamber of Commerce Cybersecurity Series, where he spoke about cyber risk and third-party risk management. [video width="1280" height="720" mp4="https://www.fico.com/blogs/wp-content/uploads/2019/06/Doug-Clare-Cyber-Video.mp4"][/video] "It’s important for consumers to realize that when they do business with somebody, they may be doing business with more than one party," Clare said. "Everybody’s got a supply chain." For consumers, protection means paying close attention. "It’s important to stay vigilant, particularly about email," Clare said. "If you get an email with a link, check it out, and don’t click it until you’re sure. Make sure the email is coming from who you think it’s coming from, that the domain name on the email address is correct. Email is a big challenge, be careful." Clare also urged viewers, "Have that conversation with your kids and your parents." Third-party risk management is a hot issue in the world of cybersecurity, since vulnerabilities in a firm's supply chain, partner or vendor networks can expose sensitive data. It's estimated that half of all data breaches occur through third parties. According to Ponemon Institute’s 2018 Data Risk in the Third-Party Ecosystem, more than 60 percent of US CISOs have indicated being the victim of a third-party breach incident. More tips on how to protect yourself online are discussed in this blog post and video. The post Cybersecurity: How to Protect Yourself Online (Video) appeared first on FICO.

Last week the Federal Communications Commission (FCC) passed a new declaratory ruling aimed at addressing rising consumer complaints about robocalls. Widely available, low-cost automated technologies have unleashed a flood of these calls. One study says the number of robocalls to US phone numbers soared from 18 billion in 2017 to more than 26 billion in 2018. There’s certainly broad support for recent efforts by the FCC as well as Congress to combat fraudulent or illegal calling — estimated to account for about a quarter of all robocalls to US mobile numbers. But the new FCC ruling, which authorizes phone companies to implement call-blocking services as a default, does not stipulate fraudulent or illegal calls. Blocking could therefore also apply to unwanted calls — a subjective categorization with unknown implications for legitimate businesses and consumers. The debt collection industry, as well as many other business sectors, will continue to urge refinements and clarifications to the ruling, and there appears to be some support from FCC Commissioner Michael O’Rielly for such changes. Here are the main challenges I see with the rule as currently written: Lack of definition and transparency about types of calls that will be blocked. Carriers are authorized to implement blocking programs “based on any reasonable analytics designed to identify unwanted calls.” Criteria and methods are likely to vary from carrier to carrier, and there may be little transparency into how blocking decisions are made. Nor is there as yet any clear mechanism for businesses to remedy erroneous blocking. Default blocking with all-or-nothing opt-out for consumers. Consumers who feel blocking is preventing them from receiving necessary calls will have to opt-out of their carriers’ service. But since consumers who do that will have no protection against illegal calls, it’s unlikely many will choose the option — making it a fairly meaningless provision. Additional consumer-controlled opt-in blocking layer. Carriers can also offer consumers an additional opt-in layer of blocking, where they can choose, for example, to block all calls except from numbers they explicitly “whitelist” or that are in their contacts list. This will only make it easier for people who regard debt collection calls as spam to avoid them altogether. Is the FCC Ruling a Problem for Your Business? Is the FCC ruling necessarily harmful to the debt collection industry, which currently largely operates through phone calls? Is it a signal, along with the Consumer Financial Protection Bureau’s newly proposed debt collection rules — limiting calls to one contact a week or seven attempts weekly while allowing email and texts — that debt collectors should switch en masse to those channels? I believe the answer to both questions is no. Shifting operational methods that have contributed to consumer discomfort with phone calls over to other channels will only reproduce the situation in those channels. (It’s estimated that only half of all cellphone calls are being answered at all—what’s to stop consumers from similarly ignoring email and texts?) Instead, the recent actions of federal agencies are a signal for debt collectors to use modern technologies to take a far lighter, more precise touch with consumers — while dramatically boosting collected amounts and operational efficiency. Case in point: A collections organization currently using this approach expects that if the CFPB proposed debt collection rules went into effect tomorrow, calling limits would affect only about one in a thousand of the consumers they contact. This is how you elevate collection calling, emailing or texting to efficient, helpful services recipients no longer perceive as “unwanted” and are likely to respond to positively. I’ll be talking about just that, along with my co-presenter Chris Cistrone, Principal Data Scientist at ConServe, at the upcoming ACA International Convention, July 14-16 in San Diego. If you’re plannin

Fraud and data breaches have always had a close, if destructive, relationship. As the US transitioned to hard-to-counterfeit EMV payment card technology several years ago, criminals flocked to card not present (CNP) fraud, often combining identity fragments and card numbers stolen in breaches to make illicit purchases online. Five years later, data breaches and downstream fraud continue their symbiotic relationship, with a steady increase in synthetic identity fraud. Financing Cars with Synthetic Identities It’s true: synthetic identities have become a major method for perpetrating auto lending fraud. Criminals are using parts of both fabricated and real identities (mined directly, or stolen during data breaches and purchased off the Dark Web) to create synthetic identities, which in turn are used to secure auto loans or other financial products. Synthetic identities can also be cultivated over time by various means, such as a legitimate cardholder getting an additional card for a person who does not exist, a process also known as “pollination.” In the case of auto fraud, once a synthetic fraudster has possession of the new vehicle, they will often ship them overseas and immediately abandon any loan payment obligations. The fact that auto lending synthetic fraud has been increasing — it is up 500% since 2011 — is an indication that many of the synthetic identities pollinated or otherwise created years ago, and cultivated to appear credit-worthy, are moving into the bust-out phase. Today’s Data Breach Is Tomorrow's Fraud I recently talked about synthetic auto loan fraud with executives from Santander Bank and GM Financial at the AFSA Vehicle Finance Conference, on a panel discussion about cybersecurity and third-part risk management (TPRM). Synthetic identity fraud provides a vivid illustration of the evolving continuum of cybersecurity and fraud: Party A’s data breach today (facilitated by poor cybersecurity defenses) becomes Party B’s synthetic identity fraud tomorrow. In the auto lending industry, most cases involve a multi-step process between one party’s data breach and another’s fraud. Sometimes these parties are business partners operating in the same automotive ecosystem. In dollar terms, synthetic loan fraud comprises about $600 million of the $1.2 trillion in outstanding auto loans. That’s a small proportion overall, but still a significant number in terms of fraud losses. An Empirical Tool to Gauge Third-Party Risk Synthetic identify fraud is a sobering outcome of the unknown, and largely uncontrolled, cyber risk exposure companies face from the partners they do business with. Addressing it requires effective third-party risk management (TPRM), starting with a baseline measurement of business partners’ cyber risk. The FICO® Cyber Risk Score is an ideal empirical tool to measure and monitor third-party risk exposure at any scale. FICO enterprise customers are using the Cyber Risk Score to continuously measure the third-party cyber risk posed by tens of thousands of partners (and more) they do business with. TPRM is a big theme for FICO and the entire enterprise cybersecurity industry, because companies recognize that while their own cybersecurity defenses may be strong, those of the third and fourth parties (vendors of vendors) they connect with may not. PwC, which, along with Deloitte, KPMG and McKinsey, has a major TPRM practice, sums up the business imperative: “In a business landscape loaded with potential pitfalls like cyber threats … that result in supply chain disruption, making sure your partners are following appropriate procedures is vital and will enable you to avoid risks and reputation damage.” Using the FICO Cyber Risk Score to empirically assess third-party cyber risk is a critical first step. In addition to helping organizations recognize and measure cyber security risk, for themselves and for their extended supply chain, FICO is an industry leader in fraud detection and prevention technologies. For m

Hope you had a relaxing Easter weekend with family and friends! In between coloring eggs, if you picked up a paper or scrolled through a news site, you probably would have seen the news that $130 million cash was found at the house of the ousted president of Sudan and how the US government foiled a money laundering operation by drug traffickers using bitcoins. Money never sleeps, as they say. Not even for Easter and such events are on the rise and a cause of constant worry for compliance officers all around the world. AI AML Models: Solving a Big Problem So, here’s the key problem. Regulatory bodies around the world have pegged global money laundering at $2 trillion. What’s worse, they estimate that less than 1% is ever caught. So to try and combat this regulators have raised the penalties for institutions found to have facilitated money laundering whether it was unknowingly or not. This has seen total fines on global financial institutions skyrocket to $27 billion in the last 10 years with Asia Pacific accounting for $600 million of that. To cope, financial institutions have hired armies of compliance officers and the median salaries in this segment have increased by 30% YOY. Lexis Nexis reports that the average cost of compliance for financial institutions in APAC is as high as $1.5 billion. At the same time, the alert volumes from existing compliance systems have increased, but still, less than 1 in 100 alerts get converted to Suspicious Activity Reports. Citing a report from ACAMS, one of the top global banks embroiled in investigations for money laundering related to drug traffickers was struggling with approximately 3,800 alerts per FTE per day. As a result, banks have large teams of talented individuals who must work through piles of false alerts to effectively find a needle in the haystack! AI AML Models: Reducing 90% of False Positives Our Chief Analytic Officer at FICO, Dr. Scott Zoldi started to examine this problem a few years ago and saw an opportunity. Leveraging the strong analytics experience in fighting fraud over 25 years, FICO announced the availability of advanced analytics to combat this problem (Using AI and Machine Learning to Improve AML). With patented techniques from fraud analytics like collaborative profiling and outlier analysis, early results indicated a reduction in false positives of up to 90% while catching 50% more suspicious transactions previously undetected. Like FICO Falcon analytics, we found a significant separation between frauds and false positives at higher score bands. In a Proof of Concept undertaken with a top 10 European bank, 42% of the highest scoring transactions consisted of only 10 false positives for every 1000 alerts generated above the score cutoff at the customer level. These models leverage two exciting new technologies. First, these unsupervised models use soft clustering to check misalignment within clusters based on behavior patterns. This ensures that the model stays current and adapts with changing behavior patterns. Second, it uses Explainable AI – top reasons associated with the score are attached which not only satisfies the regulators but also makes the cases easy to investigate. A detailed description of the concepts can be found at Deep Dive: How to Make “Black Box” Neural Networks Explainable. AI AML Models: Adding Scores To What You Have These models can be deployed as a standalone component to supplement existing rules based anti-money laundering systems to enrich the data with scores that could be used for both enhanced detection and for alert prioritization. This capability leveraging the FICO Decision Management Platform can be invoked as a microservices call and is being made available for both on premise and cloud. So for all those banks in APAC whose compliance teams have come back from the Easter break to a mountain of alerts, I am sure sorting through these false positives is causing more indigestion than the chocolate they had. But there is an eff

After a 2018 that had its highs and lows, what might 2019 have in store from a credit risk management standpoint?  Here are four key developments in credit scoring that we will be keeping an eye on in the new year: Consumer-Contributed Data Takes Center Stage Momentum is high in the consumer-contributed data space: consumers are getting more comfortable with sharing their data, provided they are presented with clear benefits for doing so.  Mandates, such as the Revised Payment Services Directive (PSD2), are ushering in the era of Open Banking around the globe.  Additionally, developments, such as the recent launch of the Financial Data Exchange (FDX), point to increasing collaboration between financial institutions and data aggregation vendors (such as Finicity, Plaid, Quovo, and Envestnet | Yodlee) to facilitate secure and efficient transfer of consumer-permissioned financial data. In 2019, enhanced credit underwriting via digitally contributed-consumer data will hit the mainstream.  With solutions such as the recently announced UltraFICO™ Score, lenders will be able to efficiently access and use verified data that reflects responsible financial activity to gain deeper insights into the credit risk profile of prospective customers. This will enable lenders to more effectively match the best credit offer to consumers. The increasing availability of solutions that utilize consumer contributed data, such as UltraFICO™ Score, will help to further empower consumers to obtain the credit they seek under competitive terms, particularly for those with sparse or inactive traditional credit files.  FICO research has found that 7 out of 10 consumers who exhibit responsible financial behavior in their checking and savings accounts could see a higher credit score with the UltraFICO™ Score. Risk in Bankcard Originations on the Rise Since 2015-2016, we have observed a shift downwards in the relationship between repayment odds (defined as the number of on-time payers for every one defaulter) at a given FICO® Score in the U.S.  This shift has been most notable for the bankcard originations population.   Often, a downward shift in the odds-to-score relationship leads to tightening of underwriting, as lenders seek to ensure that new bookings are appropriately aligned with their risk tolerance.  There is some evidence that this tightening is occurring, with the Federal Reserve Board (Fed) reporting higher reject rates based on its Oct. 2018 SCE Credit Access Survey. Per figure 1 below, the shift in odds-to-score for bankcard originations has generally been a parallel one, indicating a systemic shift in risk across the board.  Put another way, across all FICO Score ranges, the likelihood that a consumer with a newly issued bankcard will pay that card as agreed has decreased.  The parallel shift implies that the ability of the FICO Score to rank-order repayment risk remains high: instead, there is a factor impacting new bankcard repayment beyond what is captured in the credit report.  No clear explanation for this downward shift in repayment odds has emerged---our research has found that this shift is consistent regardless of factors such as geographic region, age, and card payment behavior (e.g., revolver vs. transactor). We will be closely monitoring the odds-to-score relationship in 2019.  It will be interesting to see if lender efforts to stem this trend via changes to their underwriting policies and the use of new data sources will offset further repayment pressures that are emerging, such as rising interest rates and increased volatility in the financial markets. Average FICO Score — Has It Peaked? Speaking of emerging pressures, will 2019 be the year that the streak of 8+ consecutive years of increases in the average national FICO® Score comes to an end?  Since October 2009, the average year-over-year FICO Score has steadily and consistently increased, from a low of 686 in 2009 to the latest high of 704 as of 2018. This has been driven by a

As advanced analytics permeated nearly every industry in 2018, FICO’s thought leaders continued to push it into new areas. Here were the top 5 posts in the Analytics & Optimization category last year. Advancing the Field of Prescriptive Analytics Mathematical optimization, or prescriptive analytics, applies robust science to decision strategies in order to reach the best outcome. Horia Tipi noted a breakthrough for operations researchers and data scientists worldwide: Last week we announced that FICO Xpress Mosel, the leading optimization modeling, analytic orchestration, and programming language is now open and available to everyone free of charge. From the boardroom to the classroom, anyone can now create optimization models to solve problems more efficiently and make better business decisions based on data. FICO Xpress Mosel is available by downloading the FICO Xpress Community License. With FICO Xpress Mosel, organizations can create optimization models that can solve bigger problems more efficiently, design solutions faster, and make better decisions in virtually any business scenario. In addition to its modelling, solving and programming features, FICO Xpress Mosel also supports the orchestration and execution of analytic models built in virtually any tool. Whether a problem needs solving in milliseconds, requires a vast array of cloud computing resources, or has to solve for hundreds of millions of decision variables, Mosel is there to meet the challenge. For example, Southwest Airlines have been using FICO Xpress, including Xpress Mosel, for years to handle some of their biggest, most critical business problems. Read the full pos The Financial Industry’s Digital Transformation How are analytics helping banks accelerate their digital transformation, while keeping the customers at the center of their strategies? Manish Pathak explored a roadmap for success. Financial institutions understand the need to tailor experiences to individual needs and personalize their interactions. In fact, more than half (55%) of bankers plan to increase spending on customer experience initiatives [CSI], and nearly 80% consider it important to deliver guidance to customers in real-time [The Financial Brand]. Currently, though, only about 20% of financial institutions are delivering more than basic personalization [Digital Banking Report/Everage]; clearly, there is still a significant gap to fill. We’ve identified three key imperatives financial institutions need to address to deliver personalized, real-time experiences. #1 Focus on Data Gathering and Consolidation All of these multichannel, “always on” interactions that we’ve referenced generate large amounts of data, which is typically captured in various sources such as CRM applications, transactional data stores, disparate account management data stores, etc. Each data source provides a fragmented image of a consumer—a glimpse into one of the personas. Financial institutions need to bring these data sources together to create a comprehensive profile of a consumer, rather than a series of disconnected account holders across platforms and lines of business. By connecting the digital clues and gaining a single customer view, it’s then possible to interpret and anticipate future needs. Currently, according to Forrester only 0.5% of all generated data is analyzed. Richard Joyce, a Senior Analyst at Forrester says, “Just a 10% increase in data accessibility will result in more than $65 million additional net income for a typical Fortune 1000 company.” These stats are simply overwhelming and identify a clear-cut target for which the industry must aim. #2 Build Powerful Analytic Engines that Predict and Prescribe Personalization is not a matter of simply gathering data, but also acting on that data. And the hard truth is that anticipating customer needs requires powerful analytics engines that were once thought of as “nice-to-have.” However, only about 55% of organizations were expected to in

U.S. Government Predictions 2019: Now that we have completed the 2018 election season, people are asking what is in store for 2019.  The good news is that revenues look strong, but there are a number of factors if you look deeper into the numbers.  Also, while often the first year of a legislative session occurs before our political leaders are thinking about re-elections, election cycles have apparently gotten longer, because for many American politicians, the 2020 election season has already begun. Tight Budgets Despite Increases In Overall Spending While the National Association of State Budget Officers has reported a healthy 4.3% increase in general fund spending, most of that new money is going to three places: Medicaid, K-12 education, and pension obligations.  This will likely squeeze out revenue increases in virtually all other areas.  Government will carefully scrutinize any new spending, and will be looking for proverbial “singles”, rather than “home runs”.  They will look for opportunities to quickly implement small improvements rather than embarking down multi-year, multi-million dollar projects. Within this context, government will be looking for opportunities to enhance the IT systems they already have.  They will likely look favorably at new capabilities that increase their prior investments and get additional functionality without major new initiatives.  This will include looking towards analytics as a way to deliver government services in a more efficient manner.  Through the use of optimization tools, government can identify opportunities to enhance customer service and performance outcomes without increasing staff requirements. Continued Pension Challenges Drives Government HR Policies In what may be one of the most under-reported government news stories of the year, Moody’s reports unfunded state pension liabilities have now increased to $1.6 trillion, up from $1.3 trillion just a year before.  To put this in perspective, every American would have to give approximately $4,800 to cover this deficit.  Another way to look at it is that if Illinois abandoned all state funding (police, K-12 education, prisons, etc.) for the next six years, and took 100% of their current tax revenue and allocated it 100% to its pension fund, the state would still be underfunded for their promised payouts. Within this environment, states are going to need to look towards significant changes, including changing the structure of their plans (especially for new hires), increasing the amount of money placed into their pension funds.  In most cases, state’s cannot reduce payments to retirees, they are typically guaranteed to the retiree through the state’s constitution.  Some states have greater flexibility to reduce their Other Post-Employment Benefits (OPEB), most typically healthcare subsidies for retirees.  The existing OPEB liabilities for states are estimated to be $692 Billion, as of 2015. Within this environment, government will be looking for solutions that provide enhanced service without adding additional staff which would incur new pension liabilities.  This may include automating currently manual processes, outsourcing back-office or service functions to private businesses, or eliminating services that are deemed non-essential. Enhanced Customer Service Delivered Through Self-Service Within the context of these pension challenges, government will continue to look for solutions which reduce the dependency on government staff to provide public-facing services.  These could include novel approaches to support infrequent interactions, by automating currently manual processes. For example, look for governments to move away from traditional license renewals, which require citizens to mail applications and fees to ones with automate renewals using stored payment mechanisms.  Imagine if rather than getting a renewal form in the mail, and having to mail a check for a motor vehicle registration, the government simply emailed the ca

Since the November elections, much of the talk here in Washington, D.C. has been on preparing for a divided Congress. While many believe that the partisan divide will grow in 2019, this does not mean that all bank-related public policy proposals will be stuck in neutral. In fact, when it comes to regulatory reforms promulgated by federal agencies, there could be some significant and welcome news for financial institutions. Below, are my predictions (they’re more like educated guesses) on some important issues of interest to the banking industry. The FCC Will Issue a Major Declaratory Ruling on the Telephone Consumer Protection Act (TCPA) This past year has been one of promise and progress for those waiting on TCPA reform. In March, the U.S. Court Appeals for the District of Columbia vacated key parts of the Federal Communications Commission (FCC) 2015 TCPA Order, which had created much confusion and sparked increased litigation. This month, the FCC approved the creation of a National Reassigned Number Database to address the nearly 100,000 wireless numbers that are reassigned each day. While the database is not expected to be operational for several years, the FCC did provide for a safe harbor that will protect those that leverage the database from TCPA liability. The next step? I expect the FCC to issue a declaratory ruling in the first half of 2019 that will provide much needed clarity around the definition of Automatic Telephone Dialing System. This will lead to a favorable development for consumers who opt-in and businesses that utilize technology like FICO’s Customer Communication Services to communicate important information to their customers via their cellphones. The BCFP Will Finally Release a Proposed Rule Governing Debt Collection We have been talking about this for several years but I believe we are now just a few months away from the Bureau of Consumer Financial Protection issuing a Notice of Propose Rulemaking aimed at modifying the more than 40-year old Fair Debt Collections Practice Act. The proposed rule will focus on third-party collectors, addressing issues such as communication practices and consumer disclosures. When will this happen? The latest fall 2018 rulemaking agenda indicates that the BCFP will issue the rule by March. I believe this will slip a few months but by June 2019 the proposed regulations will finally be available for public review and comment. AML/BSA Reform Talks Will Intensify but Meaningful Changes Will Have to Wait Bank Secrecy Act/anti-money laundering (BSA/AML) regulatory reforms are top of mind for regulators and legislators. This month, a group of federal agencies including the Federal Reserve, OCC, FDIC and the Financial Crimes Enforcement Network (FinCEN) issued a joint statement which encourages banks to consider, evaluate, and responsibly implement innovative solutions to BSA/AML compliance. Members of Congress have also been focused on BSA/AML reforms. In June, House members Blaine Leutkeymeyer (R-MO) and Steve Pearce (R-NM) introduced legislation that included an increase in the thresholds for currency transaction reports (CTRs), from $10,000 to $30,000, as well as suspicious activity reports (SARS), from $5,000 to $10,000. These are two areas where financial institutions are spending billions of dollars a year on compliance and have been looking to ensure their extensive efforts are properly calibrated to provide regulators with meaningful insights. On this topic, a FICO colleague recently wrote a blog post noting that FinCEN received more than 2,000,000 SARs in 2017, and that by employing cognitive analytics FinCEN could gain new insights from these reports. Where is BSA/AML reform headed in 2019? Legislative efforts will prove challenging. While there is some agreement to address the issue of beneficial ownership, other reforms have divided Republicans and Democrats. Despite extensive, ongoing discussions of BSA/AML reforms among regulators, a recent Senate Banking Commit

Small business credit access grew in 2018 and with modifications to the Small Business Administration (SBA) rules for more streamlined processing, access is predicted to grow. However, untapped opportunities for growth remain for startups and minority owned businesses as suggested in SBA research. What are the funding options for these groups that the SBA is offering and how can your organization better serve them as well? This issue was discussed at the NAGGL 2019, National Association of Government Guaranteed Lenders, conference where hundreds of SBA lenders gathered to learn about policy changes, SBA online application submission solutions, how to create successful sales reps, and other trends in the industry impacting SBA lenders. SBA programs give businesses access to credit such as 7(a) loans for any purpose, microloans and 504 Certified Development Company (CDC) Loans for fixed-rate mortgage and equipment financing. Other programs are available to businesses to find investors, conduct R&D or to secure bonding. The programs and NAGGL event attract mostly community banks and credit unions across the US because the SBA is known for connecting businesses with their local providers. FICO attended and exhibited at the event to connect with lenders on FICO® Small Business Scoring ServiceSM (SBSSSM), a score used by the SBA to assess risk of businesses and determine funding eligibility. Linda McMahon, Administrator of the SBA, gave the opening keynote where she shared SBA updates and the economic success that these Government Guaranteed Lenders have contributed to. Small business lending was strong in 2018 and pro small business growth policies have been conducive to that growth. However, there is even more opportunity that small businesses are not realizing. Specifically, the underserved businesses such as startups, those run by minorities and veterans, and businesses in rural areas which is now a focus area of growth for the SBA according to McMahon. To change this, the SBA is improving technology solutions to better match businesses with lenders, see quicker turnaround times, and achieve higher approval rates. They created new educational resources about programs they offer for use by lenders and borrowers to understand and share all program details, requirements and terms. FICO similarly works to help community banks and credit unions understand how they can increase loans given to those underserved businesses and do so profitably. As I spoke to these lenders, they recognize that something needs to change with the traditional underwriting processes at their organization, but they have many hurdles such as getting leadership on board with decision automation, cost of technology and cost of compliance. FICO believes it’s important for lenders to understand small business scoring and how it’s different than consumer credit scoring such as a FICO® Score. Business owners often apply for personal loans to get credit for their business so personal data is used to make the decision. This becomes a problem because personal data alone isn’t predictive of business success. Information on the business credit information, and success of similar business portfolios isn't factored into the funding decision when it must be. Having a scoring model that is built on historical business portfolio data will give business owners more chances at funding because the applications are more fairly and consistently assessed. Even with limited or no business and personal credit history. To learn more about FICO can help grow small business portfolios with scoring and give more businesses access to credit, check out our on-demand webinar for an in-depth look. Or visit our website. The post How Lenders Can Better Support Small Business Growth in their Communities appeared first on FICO.

Recently, a performance announcement from a FICO competitor caused a big shakeup in the mathematical optimization community. My colleague Timo Berthold and I wrote a detailed post about MIP benchmarking on the FICO Community blog, but I thought it was worth sharing the gist of it here. At issue is how developers benchmark the performance of mathematical optimization tools, particularly mixed integer programming (MIP) solvers. The community already has a clear set of standards for MIP benchmarking (see MIPLIB2010), which have evolved over time. In the case at hand, there were clear issues with the way the competitor generated and discussed their results: When there is a test set, clearly defined as the “benchmark set”, picking subsets of instances to justify general claims about performance is a bad and misleading practice. It is particularly troubling when statements can be read as if they held for the full set and not only for a subset.  Read more about the MIPLIB2017 benchmark set below, after the bullets. Even if one was to present comparative results on a subset, those results should (a) be put into context to the results on the full set and (b) it needs to be explicitly named which instances belong to the subset.  Neither happened in this particular case. Every community has its standard measure for performance. In computational MIP, this is shifted geometric mean(1) of running times. There are a few, let’s say “minor standards”, such as node counts, and number of solved instances for example. However, no one should use a measure for comparison that is non-standard to the community, without explaining it in detail. In the case mentioned above, this was not done consistently in the majority of ongoing communications. Non-standard measures can be tricky or misleading. In this particular case, the PAR10(2) measure computes a score number, not a speedup factor, since it multiplies some of the involved values with penalty terms. Therefore, a PAR10 score cannot and must not be used to make a statement such as “solver A is x times faster than solver B”, as has happened. PAR10 is not a speed factor. In our opinion, this is a good argument to not use PAR10 for computational MIP in general, since its results do not represent a quantitative statement. Doing all of the above and publishing a result so far off official numbers, just days before a new benchmark set and results will be published, is bad practice. What Is FICO's Take on MIP Benchmarking? When we present comparisons of FICO Xpress Optimization against competitors on MIPLIB or other sets from Hans Mittelmann's benchmark site, FICO always uses the numbers presented there and will continue to do so. FICO strikes a careful balance between putting effort in to benchmarking and delivering value to customers. We believe in the strength of the mixed-integer programing community to define their standards and we see ourselves as active members of this community. We feel honored that FICO representatives were part of the MIPLIB2010 and MIPLIB2017 committees and we stand by the results of those international research and norm-defining projects. It is an exciting time for mathematical optimization, and we hope that the great community spirit can be kept up. For a fuller description, see our original post. The post MIP Benchmarking: Don’t Abuse the Standards appeared first on FICO.