Reinforcement learning and mortgage partial prepayment behavior
یادگیری تقویتی و رفتار پیش پرداخت جزیی وام -2020
An agent can learn from previous experience to make decisions. Several important studies claim that reinforcement learning plays a key role in explaining the evolution of the individual learning process. This paper studies the likelihood of mortgage partial prepayments and the process through which mortgage borrowers learn from making partial prepayment decisions in the residential mortgage market in China. Learning dynamics are measured by studying the mortgage partial prepayment behavior of individual borrowers. A longitudinal discrete choice model of the choice of the mortgage payment is presented and estimated using a rich set of mortgage loan history data from a leading mortgage lender in China. The results indicate that path dependency and reinforcement learning arise whenever a borrowers “partial prepayment” decision depends not only on current-stage variables and his/her individual characteristics but also on the learning experience (both from himself/herself and others). Borrowers with more partial prepayment experience in previous stages have a higher probability of making the same decision in the future. Moreover, learning dynamics are not monotonic, and recent experience plays a larger role than distal experience in determining a partial prepayment decision.
Keywords: Mortgage partial prepayment | Reinforcement learning | Learning by doing | Recency
An explainable AI decision-support-system to automate loan underwriting
یک سیستم پشتیبانی تصمیم گیری هوش مصنوعی قابل توضیح برای اتوماسیون پذیره نویسی وام-2020
Widespread adoption of automated decision making by artificial intelligence (AI) is witnessed due to specular advances in computation power and improvements in optimization algorithms especially in ma- chine learning (ML). Complex ML models provide good prediction accuracy; however, the opacity of ML models does not provide sufficient assurance for their adoption in the automation of lending decisions. This paper presents an explainable AI decision-support-system to automate the loan underwriting pro- cess by belief-rule-base (BRB). This system can accommodate human knowledge and can also learn from historical data by supervised learning. The hierarchical structure of BRB can accommodates factual and heuristic rules. The system can explain the chain of events leading to a decision for a loan application by the importance of an activated rule and the contribution of antecedent attributes in the rule. A business case study on automation of mortgage underwriting is demonstrated to show that the BRB system can provide a good trade-offbetween accuracy and explainability. The textual explanation produced by the activation of rules could be used as a reason for denial of a loan. The decision-making process for an application can be comprehended by the significance of rules in providing the decision and contribution of its antecedent attributes.
Keywords: Explainable artificial intelligence | Interpretable machine learning | Loan underwriting | Evidential reasoning | Belief-rule-base | Automated decision making
Insight into simultaneous catalytic oxidation of benzene and toluenein air over the nano-catalyst: Experimental and modeling viaCFD-ANN hybrid method
بینش به اکسیداسیون همزمان کاتالیزوری هوای بنزن و تولوئنین بر روی نانو کاتالیزور: آزمایش و مدل سازی از طریق روش ترکیبی CCD-ANN-2020
tThis study reveals the simultaneous deep oxidation of benzene and toluene over the novel supportedcobalt oxide catalyst derived from metal organic framework (MOF) over the almond shell based activatedcarbon. The performance of the fabricated catalyst was evaluated under the various operating conditionsincluding oxidation temperature, initial concentration of benzene and toluene. The maximum conver-sion of benzene and toluene were also measured to be 89.74 % and 82.37 %, respectively. The samplemorphology was studied by applying XRD, FESEM, BET and TGA analysis. The characterization tests indi-cated that the well dispersed spherical nano-supported catalyst was synthesized with size of less than40 nm. To the best of our knowledge, the computational fluid dynamics (CFD) analysis incorporated withartificial neural network (ANN) was also studied for modeling the deep catalytic oxidation over the pre-pared sample. The modeling involved with the three dimensional analysis of polluted air flow through ofa tubular micro-reactor axial inlet and outlet. The computational fluid dynamics was coded by adoptingCOMSOL Multiphysics to model the catalytic conversion of volatile organic compounds (VOCs) insidethe porous media. The kinetic modeling was also conducted by using three-layer ANN to determine thereaction rates while the reaction temperature, initial concentration of benzene and toluene were consid-ered as the input variables of network. The reaction rates were calculated by a non-linear feed-forwardnetwork with 5 neurons and log-sigmoid function in the hidden layer while the correlation coefficientwas achieved to be 0.99. The validation of CFD model was accomplished which showed the appropriatematching between the experimental data and model achievements. Therefore, the developed intelligenthybrid model (CFD-ANN) in the offered investigation can be a useful tool for studying the fluid dynamicsof VOCs oxidation over the nano-catalyst under the different operating conditions.
Keywords:OxidationMetal organic framework | Computational fluid dynamic | Neural network
خواص مکانیکی و دوام بلند مدت پلاستیک پلی سولفون بازیابی شده
سال انتشار: 2019 - تعداد صفحات فایل pdf انگلیسی: 11 - تعداد صفحات فایل doc فارسی: 28
در بازیابی پلاستیک های پلی سولفون، صرفه جویی خوبی در انرژی و حفاظت از محیط زیست قابل توجهی دیده می شود. این مقاله با ویژگی های مکانیکی و ماندگاری و دوام بلند – مدت پلاستکی های دست نخورده و بازیابی شده پلی سولفون که از ضایعات غیربافتی پلاستیک های پلی سولفون جمع آوری شده اند ارتباط دارد، بررسی مکانیکی آزمایش کشش و آزمایش تاثیر ایزود برای بررسی تاثیر فرآوری چرخه ای روی عملکرد پلاستیک های پلی سولفون انجام می شوند. دوام بلند – مدت پلاستیک های دست نخورده و پلی سولفون برمبنای انطباق زمان – دما و با استفاده از یک تحلیل مکانیکی دینامیک (DMA) مطالعه می شود. پایداری حرارتی ازطریق انرژی فعالسازی پیرولیتیک محاسبه شده توسط روش سنتیکی تبدیلی ایزو و با استفاده از یک تحلیل حرارتی – گرانشی (TGA) بررسی می شود. نتایج نشان می دهند که پلاستیک های بازیابی شده پلی سولفون خواص کششی مشابه و درعین حال استحکام ضربه ای پایین تری را نسبت به پلاستیک های پلی سولفون دست نخورده نشان می دهند. دوام بلند – مدت و پایداری حرارتی پلاستیک های دست نخورده نسبت به پلاستیک های بازیابی شده بهتر است و با افزایش زمان فرآوری چرخه ای کاهش می یابد، که به ترکیب کردن ناخالصی ها و تجزیه ساختار مولکولی در فرآیند بازیابی نسبت داده می شود.
|مقاله ترجمه شده|
Household debt, financial intermediation, and monetary policy
بدهی خانوار ، واسطه گری مالی و سیاست های پولی-2019
The collapse of housing prices in the U.S. during the Great Recession eroded not only consumers’ housing wealth but also the assets held by the banking sector. I introduce a micro-founded banking sector to a standard DSGE model with household debt to study the interaction between housing prices, household debt, and banks’ balance sheet positions. I estimate the model using US data from 1991Q1 to 2014Q1 and find that there is a significant spillover effect from the housing market to the rest of the aggregate economy. The spillover effect is mainly evident on investment through the banking sector. A negative shock to housing demand or to the perceived riskiness of assets backed by housing wealth decreases the banks’ net worth. As a result, both mortgage and corporate spreads rise, leading to a decline in aggregate investment. I also find that an unconventional monetary policy is more effective in dampening the downturn when it targets the assets backed by housing wealth.
Keywords: Household debt | Mortgage spread | Banking | Unconventional monetary policy
Modeling Diversification and Spillovers of Loan Portfolios’ Losses by LHP Approximation and Copula
تنوع مدل سازی و سرریز از تلفات پرتفوی وام توسط تقریب و کوپلای LHP-2019
This paper suggests a top-down method for aggregating the economic capital of an entire banking system and decomposing it into loan sectors according to their risk contributions. We model the individual losses of loan sectors by large homogeneous portfolio (LHP) approximation based on multi-factor skew normal credit worthiness and combine them by applying static and dynamic copulas to reflect diversification effects and spillovers across loan sectors. Our method is more efficient and practically useful than typical multi-factor models using numerical integration due to the latency of risk factors in that losses are directly generated by Monte Carlo simulation using copula without knowing any risk factors. As a result of our empirical study on charge-off rates of the U.S. commercial banking system, we find that the residential real estate loan sector is the most affecting as its default risk spills over to the rest of the banking system, and hence its risk contribution to the entire banking system is large. However, the commercial real estate loan and business loan sectors are revealed to be affected sectors whose risk contributions are large, but the losses are mainly due not only to their large exposure size, but also to default contagion from others. The risk contributions of credit cards and other consumer loans as default risk affecting sectors become larger in terms of the recent conditional dependence. Lastly, using time-varying correlation analysis, we find that the subprime mortgage crisis is a systemic event that affects the entirebanking- system, while the commercial real estate and the dotcom bubble crises are sector-wide systemic events.
Keywords: Multi-factor mode | Copula | Loss distribution | Diversification | Spillover
Tougher than the rest? The resilience of specialized ﬁnancial intermediation to macroeconomic shocks
سخت تر از بقیه؟ تاب آوری واسطه گری مالی ویژه در برابر شوک های کلان اقتصادی-2019
This paper uses a unique and comprehensive dataset comprised of 41 years of detailed banking datafrom the German building society industry to provide empirical evidence of the resilience of specializedfinancial intermediation (SFI) to macroeconomic shocks. Compared to studies on the German bankingindustry, we show that SFI is extremely stable despite generally less diversified revenue streams. We arealso able to illustrate the impact of various economic conditions and cycles on the interactions betweenthe building society sector and macroeconomic development. We use a VAR approach to demonstratethat 1) macroeconomic and mortgage-specific market shocks have limited or no impact on the bankingperformance (return on equity) or distress (write-off) indicators, and 2) the results of previous studiesshowing that financial rewards lead to higher bank funding stability on a contractual level also hold whenanalyzing data on an institutional level. As much research has found, and as the recent worldwide finan-cial crisis highlighted spectacularly, banking stability is fundamental to the overall stability of financialsystems and economies as a whole. Therefore, our results contribute to the ongoing discussion of whetherspecialized financial intermediation can indeed lead to more stable banking systems.
Keywords:Stress testing | Macroeconomic shock | Financial intermediation | Financial stability | Building societies | Savings and loan contracts | VARa
Effects of government bailouts on mortgage modification
تاثیرات فرار دولت روی اصلاح رهن گذاری-2018
This paper shows how liquidity infusions affect loan modification in the mortgage market. The design of pooling and servicing agreements leads mortgage servicers to prefer foreclosure over modification when they are liquidity constrained. Therefore, a positive liquidity shock is expected to boost modification rates. Using a residential mortgage dataset that includes loan-level information, we find that the Troubled Asset Relief Program significantly increased the modification rate. Our findings help us better understand the economic consequences of government intervention and have important policy implications for the renegotiation of distressed mortgages.
keywords: Mortgage modification |Financial crisis |TARP |Government intervention |Liquidity
Sellers optimal replenishment policy and payment term among advance, cash, and credit payments
سیاست تجدید تدارکات بهینه فروشنده و بخش پرداخت دربین پرداخت های پیشرفته، نقدی و اعتباری-2018
It is evident that granting a short-term interest-free loan (i.e., a credit payment) stimulates more sales than asking for an advance payment. In addition, it is obvious that a 30-year mortgage has a higher default risk than a 15-year mortgage. As a result, it can be inferred that the longer the credit period, the higher the sales volume as well as the higher the default risk. Conversely, there are no default risks with an advance payment. Also, the longer the prepayment period (i.e., advance period), the lower the sales volume but the higher the interest earned. In this paper we incorporate the above mentioned important and relevant phenomena into the proposed model. Hence, the payment period and the replenishment cycle time are decision variables. We then derive the sellers profit under each of the three payment terms: advance payment, cash payment, and credit payment. In addition, we obtain explicit closed-form solutions to the problem, and explain them by simple economic interpretations. Furthermore, we demonstrate that an increase in selling price elevates the payment period, while an increase in purchasing cost reduces the payment period. Finally, we perform sensitivity analyses to examine the impacts of financial related parameters on the sellers decisions and profits, and then provide several managerial insights. For example, if the impact of advance payment on demand is relatively smaller than that of credit payment, then it is more profitable for the seller to ask for an advance payment than to offer a credit payment, and vice versa.
keywords: Economic order quantity |Finance |Permissible delay in payment |Advance payment |Cash on delivery
House price, loan-to-value ratio and credit risk
قیمت خانه، نسبت وام به ارزش و خطر اعتباری-2018
Real estate transactions are often established through financing. We study the effect of financing on property prices. We show that properties can transact at prices well above their collateral values. Therefore, the commonly used loan-to-value (LTV) ratio suffers a bias that can significantly understate credit risk. This bias is exacerbated when mortgages are originated with longer terms, at higher LTV ratios, or when sellers possess stronger bargaining power. Furthermore, this bias is larger under aggressive lending products, e.g. interest-only loans and mortgages allowing negative amortization. Our simulation results suggest that many mortgages originated at the peak of the housing bubble are, in fact, “under water” at origination. In particular, the loan amount of a 30-year mortgage at a 95% LTV can be 15% greater than the collateral value of the property, suggesting the mortgage is already deep “under water” at origination. These findings call into questions underwriting and risk control practices in mortgages and other collateralized debts.
keywords: Asset prices |Bargaining |Mortgage financing |Loan-to-value ratio |Credit risk