About KDS Global

Our UBiquitous indeX (UBXTM) matrix computing architecture based artificial intelligence & machine learning technology infrastructure is used to manage the 500TB BIG mortgage loan-level and pool-level data to generate predictive competing risk scores to provide daily MBS pricing analytics for the 1.5M MBS CUSIPs. Anyone who works on the sell-side, issuers, buy-side, portfolio managers, risk manager, traders, originators, servicers can all use our UBXTM capabilities and capacities to vertically manage their MBS execution strategies and critical real-time trading.

Click HERE to review our products and services in details.

Some Cool Facts

500

Terabytes of Mortgage Data

250,000

Lines Coded in UBX System™

100

Happy Clients

U.S. Patent Applications

UBX Virtual Pocket Sorting & Indexing

The present invention sorts very large volumes of data records by forming successive lists of sorted record start address by character rank from a LSD (least significant digit) of a sort field to a MSD (most significatn digit) of such field with each list being formed in the order of the preceeding list...

QED Options Pricing Model

In the financial world, option trading makes up a large part of the opportunities to make money. Determining the future market value of these securities and options in an efficient manner is important to all the entities involved buying, selling, and holding these options and securities. Traders, trustees, investment bankers, fund managers, broker-dealers ...

Unified Differential Economics

Capital markets and global economic behavior have long been studied as disjoint sets of events. In particular, within the broad fields of securities pricing, risk management, and financial economics, the daily (and intra-daily) phenomena that define these fields are analyzed by respective market participants under a plethora of methods. Such methods range in both complexity and fundamental understanding ...

Available TBA-Only MBS Supply (ATOMS)

In the financial world, home loans are usually organized, optimized by groups and structured into mortgage backed securities (MBS). These MBS are bought and sold on the bond market. Determining the market value of these securities in a highly transparent and efficient manner is important to all the entities involved in creating, buying, selling, servicing, and holding these securities ...

UBiquitous IndeX (UBXTM)

Typical data analysis involves ad hoc query, summation, sorting, and linear table scan. The organization of data and metadata greatly affects the performance and ease of use. When performing ad hoc queries, many existing analytical tools use a combination of indexing of some frequently used fields and linear table scan on non-indexed fields ...

Deconvolution Risk Scores

In the financial world, home loans are usually organized, optimized by groups and structured into mortgage backed securities (MBS). These MBS are bought and sold on the bond market. Determining the market value of these securities in a highly transparent and efficient manner is important to all the entities involved in creating, buying, selling, servicing, and holding these securities ...

Unified TSP Index Return and Price Algorithms

In the US financial markets residential loans are typically organized by borrower rate, rate type, and loan term for the purpose of securitization into mortgage backed securities (MBS). In addition, certain MBS, similar in rate, rate type, and loan term are deemed “more desirable” than others. These mortgage(s) or mortgage pools are commonly referred to as “specified pools.” ...

Risk-Adjustd Yield (RAY) MBS Pricing

Traditional pricing of Mortgage Backed Securities utilizes collateral level data as well as prepayment and default projections to derive cash flows, which are then discounted at a prevailing market rate that is often opaque across buyers and sellers ...

Unified ATOMS Index Return and Price Algorithms

Home mortgages are usually organized, optimized by groups and structured into mortgage backed securities (MBS). These MBS are bought and sold on the bond market. Determining the returns from these securities in a highly transparent and efficient manner is important to all the entities involved in creating, buying, selling, servicing ....

Machine Learning Algorithm for Personal Portfolio Wealth Creation

In the 7x24 global trading market environment, it is critical for an investor to have access to the BIG DATA about the past, present and future cash flow value added or risk-adjusted prices and anticipated future changes in market ask vs bid transaction prices automatically. To this end, models are developed that attempt to predict the future ...

CMO Cashflow Rule-Base Library Builder

A CMO (Collateralized Mortgage Obligation) is a type of mortgage backed securities that consists of multiple collateral groups, the cashflow from which are allocated to a set of tranches based on certain payment rules ...

UBiquitous linear sorted order indeXing engine (UBX)

In the time of big data there is a need to find and access the data rapidly. When performing ad hoc queries, many existing analytical tools use a combination of indexing of some frequently used fields and linear table scan on non-indexed fields. Traditional indexes take a relatively long time to generate and a relatively long time to use. Often the index for a field is bigger than the actual data field that is indexed ...

Our Clients

  • 中国国家外汇管理局 (State Administration of Foreign Exchange)
  • FEDERAL RESERVE BANK OF NEW YORK (纽约联邦储备银行)
  • J.P. MORGAN (摩根大通)
  • WELLS FARGO (富国银行)
  • STATE STREET (美国道富银行)
  • BANK OF AMERICA (美国银行)
  • CITI BANK (花旗银行)
  • CREDIT SUISSE (瑞士信贷银行)
  • GOLDMAN SACHS (高盛集团)
  • FREDDIE MAC (房地美)
  • BLACK ROCK (黑石)
  • STIFEL (斯迪富金融管理)
  • MORGAN STANLEY (摩根斯坦利)
  • LOOMIS SAYLES (卢米斯赛勒斯)
  • BNP PARIBAS (法国巴黎银行)
  • DEUTSCHE BANK (德国银行)
  • ANNALY CAPITAL MANAGEMENT (安纳利资本管理)
  • JANUS CAPITAL GROUP (骏利资产管理集团)
  • VOYA (荷兰国际集团)
  • BAYVIEW ASSET MANAGEMENT (湾景资本管理)