Category Financial Econometrics and Empirical Market Microstructure

Price-Volume Relationship

The price-volume relationship is one of the most studied in the field of finance when studying price dynamics. One of the oldest models used to study price-volume relationship is the model of Osborne (1959) who models the price as a diffusion process with its variance dependent on the quantity of transaction at that particular moment. Subsequent relevant work can be found in Karpoff (1987), Gallant et al. (1992), Bollerslev and Jubinski (1999), Lo and Wang (2002), and Sun (2003). In general this line of research studies the relationship between volume and some measure of variability of the stock price (e. g., the absolute deviation, the volatility, etc.)...

Read More

Outliers as Early Warning Signals

Outliers play a crucial role in early warning. Outliers are the first visible signal of a regime shift into abnormal markets as Early Adopters act on disruptive information. HSBC’s February 23, 2007 $10.5 bn subprime loss announcement caused a tripling of AAA subprime spreads in a single day (a 12 standard deviation outlier). Four days later, this disruptive information cascaded into broad equity markets as February 27 saw exceptional downside outliers from China to the U. S. The Dow Jones recording its 6th biggest daily surprise in over 100 years (Finger 2008) on that day. A —3.3 % drop would hardly appear noteworthy, except that it happened just as volatility reached historical lows and therefore represented a 7.8 standard deviation outlier. See Table 1 for the top 10 DJIA surprise s...

Read More


The model simulates the continuous double auction, a market structure common to most modern exchanges. Traders submit bids and offers to buy and sell respectively at the best price they are willing to transact at. If prices cross—a bid meets or exceeds a previous offer, or the converse—a transaction takes place at the earlier listed price. If an incoming order is unable to transact with any existing orders, it is placed in the limit order book. This consists of two lists, the bid book and the ask book which contain the previously unfilled orders on the buy and sell side respectively.

At each time step in our model, a random order of unit size is generated...

Read More

Scheinkman and LeBaron Test for Predictability

Подпись: CM+1 (") CM (e)
Подпись: SM+1 (e)

Another interesting use of correlation integral was presented by Scheinkman and Lebaron (1989). As before, Cm(e) stands for the correlation integral for M as a phase dimension of reconstructed space, and threshold e. It is proven that

gives an estimate of conditional probability that

sup jyi+i – У2+І |< e,

0<i <M

given that

sup |yi+i – У2+І |< e,

0 <i <M-1

Table 1

Scheinkman-LeBaron function behavior i. e. the conditional probability that two states of the system are close, given that their past M histories are close.

This result can be implemented to define the measure of predictability and determinism of the data. If SM(e) does not saturate as M grows, then states of the system depend on the information about its history...

Read More

Models and Data for Bank Ratings

Here, and further in this section, ordered probit/logit econometric models were used to forecast rating grades (for example, see Peresetsky and Karminsky (2011)). Numeric scales for ratings were also used as a result of the mappings mentioned in Fig. 1. For the main international CRAs, nearly 18 corporate rating grades were used.

The original databases for different classes of entity were used. There were two different databases used separately for banks for both international and Russian ones. The first database was obtained from Bloomberg data during the period 1995— 2009. The database includes 5,600+ estimations for 551 banks from 86 countries. The data contains the banks from different countries including more than 50 % from developed and 30 % from developing countries...

Read More

Introducing StressGrades™

Outlier based early warning is especially useful when considering the many scenarios risk managers shock their positions with. A risk manager at a global bank explained that they run close to 200 daily scenarios against hedge fund counterparties alone, but that the amount of data was overwhelming and therefore largely ignored.

This insight gave rise to the StressGrades methodology to prioritize attention on escalating market based early warning signals. StressGrades are designed to complement the existing stress testing process by (a) drawing attention on escalating visible risk, as well as (b) highlighting abnormally low visible risk themes to search for hidden risk.

9 We Define Three Volatility Based Metrics

1. PStress = Market Implied Probability of a Stress Scenario, in bps per annum


Read More