Market Predictability - page 6

New comment
 

market_depth_and_order_size.pdf

In this paper we measure market depth by investigating the relation between net order flow and price changes. Two aspects are our main focus. Is the relation linear? Is the relation different for positive and negative net order flow? Answers to these questions are important for the design of market liquidity studies and for optimal trading. We use intraday data on German index futures. Our analysis based on a neural network model provides us with two main results. First, the relation between net order flow and price changes is strongly non-linear. Large orders lead to relatively small price changes whereas small orders lead to relatively large price changes. We provide an example which shows that the optimal trading strategy of informed investors depends crucially on whether the price impact is linear or not. Second, we find that buyer initiated trades lead to a smaller price change than seller initiated trades of the same size. This finding contradicts the assumption of a symmetric price impact of buy and seller orders which is commonly used in theoretical models. Overall, the results of our paper suggest that the assumption of a linear and symmetric impact of orders on prices is highly questionable. Thus, market depth cannot be described sufficiently by a single number. Therefore, empirical studies comparing liquidity of markets should be based on the whole price function instead of a simple ratio. A promising avenue of further theoretical research might be to allow the priceimpact per unit to depend on the trade volume. This should lead to quite different trading strategies as in traditional models.
 

We document using the ZEW panel of German stock market forecasters that weak forecasters tend to be overconfident in the sense that they provide extreme forecasts and their confidence intervals are less likely to contain eventual realizations. Moderate filters based on forecast accuracy over short rolling windows are somewhat successful in improving predictability. While poor performance can be due to various factors, a filter based on a prior tendency to provide extreme forecasts also improves predictability.

 
seekers:

market_depth_and_order_size.pdf

Interesting

Anybody using NN in forex?

 

Hi Seekers

thanks for to all of your efforts,do you think this is also a kind of logical predictability,have a look at under given site.

"http://www.forexearlywarning.com/forex-lessons/parallel-and-inverse-analysis"

Hi techmac

plz review the above mentioned site,i think it is more close to real forecasting.

regards

notice ... it is not and never a kind of ad but i think it give traders a genuine reason why and which pair to trade.

 

Hi

Small contribution from my side - I found this book very interesting - http://www.amazon.com/Why-Stock-Markets-Crash-Financial/dp/0691118507

 
conclusion of this paper : "if" there's a holy grail it can't guarantee a profit at all
Files:
SSRN-id2711214.pdf  738 kb
 
This paper attempts to identify implicit exchange rate regimes for the Yen/Dollar exchange rate. To that end, we apply a sequential procedure that considers both the dynamics of exchange rates and central bank interventions to data covering the period from 1971 to 2003. Our results would suggest that implicit bands existed in two subperiods: April-December 1980 and March-December 1987, the latter coinciding with the Louvre Accord. Furthermore, the study of the credibility of such implicit bands indicates the high degree of confidence attributed by economic agents to the evolution of the Yen/Dollar exchange rate within the detected implicit band rate, thus lending further support to the relevance of such implicit bands.
 
All right. The market is absolutely predictable. After up is down, down is up after. And so constantly)))
 
Under minimal assumptions finite sample confidence bands for quantile regression models can be constructed. These confidence bands are based on the "conditional pivotal property" of estimating equations that quantile regression methods aim to solve and will provide valid finite sample inference for both linear and nonlinear quantile models regardless of whether the covariates are endogenous or exogenous. The confidence regions can be computed using MCMC, and confidence bounds for single parameters of interest can be computed through a simple combination of optimization and search algorithms. We illustrate the finite sample procedure through a brief simulation study and two empirical examples: estimating a heterogeneous demand elasticity and estimating heterogeneous returns to schooling. In all cases, we find pronounced differences between confidence regions formed using the usual asymptotics and confidence regions formed using the finite sample procedure in cases where the usual asymptotics are suspect, such as inference about tail quantiles or inference when identification is partial or weak. The evidence strongly suggests that the finite sample methods may usefully complement existing inference methods for quantile regression when the standard assumptions fail or are suspect.
12345678910111213
New comment