Probability Of Price Change Research Articles

Online retailers’ price rigidity factors and structural breaks

Modern information technologies make it possible to aggregate and study a significant amount of data on the prices of online retailers. Online data has the advantages of high frequency and real-time monitoring. Based on daily granular micro data on prices since September 2020 to May 2023 (including), we test theoretical and empirical drivers of price stickiness using a logistic regression approach. The heterogeneity of the data leads us to conduct estimates on two sub-periods: before and after February 2022. In the first sub-period we found that the use of attractive prices by online retailers contributed to increased price rigidity, while the probability of price change was higher as the more productscompetitors adjusted prices. Price changes ranged from one to four weeks. In the second sub-period, the factors of market size and attractive prices became insignificant for the pricing of online retailers. Price changes continued to be periodic and dependent on the behavior of competitors but to a lesser extent than in 2020—2021.

Inflation gap persistence, indeterminacy, and monetary policy

Empirical literature has documented that the persistence of the gap between inflation and its trend declined after the Volcker disinflation. Previous studies into the source of the decline in inflation gap persistence have offered competing views while sidestepping the possibility of equilibrium indeterminacy. We examine the source of the decline by estimating a medium-scale dynamic stochastic general equilibrium model using Bayesian methods that allow for indeterminacy. We show that the Fed's change from a passive to an active policy response to the inflation gap or a decrease in firms' probability of price change can fully account for the decline in inflation gap persistence by ruling out indeterminacy that induces persistent economic dynamics.

Price Rigidity Analysis: Evidence from the E-Commerce Market

The article studies price rigidity for food, non-food products and services based on daily data of online retailers in Moscow collected for the period from February 2019 to May 2020. It was found that the average price invariance period was 19.5 days. In addition, there is significant heterogeneity in the frequency of price changes: food prices change more often than non-food ones, and the cost of services changes the least often. The observed price behavior confirms the implications of time- and state-dependent models. The dependence on time is indicated, for example, by the fact that the probability of price changes sharply increases by days 335 and 353 of their invariability. This fact reinforces the conclusions on the presence of seasonality in price changes found in a number of foreign studies. The analysis also suggests that, in March–May 2020, growth in uncertainty caused by the coronavirus pandemic led to firms beginning to change prices more often and by a smaller magnitude. As a result, the share of price increases went up from 50.2% in March–May 2019 to 62.7% in March–May 2020. The number of days in which prices for any goods and services increased rose from 65 to 84 (out of 92 days in March–May). Thus, the results of this paper confirm that pricing behavior on the e-commerce market can vary significantly depending on economic conditions.

Trading by estimating the quantized forward distribution

ABSTRACTIn this article, a novel algorithm is developed for electronic trading on financial time series. The new method uses quantization and volatility information together with feedforward neural networks for achieving high-frequency trading (HFT). The proposed procedures are based on estimating the Forward Conditional Probability Distribution (FCPD) of the quantized return values. From past samples, the conditional expected value can be learned, from which FCPD can be obtained by using a special encoding scheme. Based on this estimation, a trading signal is triggered if the probability of price change becomes significant as measured by a quadratic criterion. Due to the encoding scheme and quantization, the complexity of learning and estimation has been reduced for HFT. Extensive numerical analysis has been performed on financial time series and the new method has proven to be profitable on mid-prices. In order to beat the secondary effects, we focus on the most liquid assets, on which we managed to achieve positive profits.

Low Complexity Algorithmic Trading by Feedforward Neural Networks

In this paper, novel neural based algorithms are developed for electronic trading on financial time series. The proposed method is estimation based and trading actions are carried out after estimating the forward conditional probability distribution. The main idea is to introduce special encoding schemes on the observed prices in order to obtain an efficient estimation of the forward conditional probability distribution performed by a feedforward neural network. Based on these estimations, a trading signal is launched if the probability of price change becomes significant which is measured by a quadratic criterion. The performance analysis of our method tested on historical time series (NASDAQ/NYSE stocks) has demonstrated that the algorithm is profitable. As far as high frequency trading is concerned, the algorithm lends itself to GPU implementation, which can considerably increase its performance when time frames become shorter and the computational time tends to be the critical aspect of the algorithm.

Predicting stock price movements: an ordered probit analysis on the Australian Securities Exchange

Using stocks from a wide range of industry sectors on the Australian Securities Exchange, this paper examines the conditional distribution of intra-day stock prices and predicts the direction of the next price change in an ordered-probit-GARCH framework that accounts for the discreteness of prices. The analysis also incorporates the endogeneity of the time between trades in an ACD framework. Other elements considered include depth, trade imbalance, and volume. The results show that trade imbalance has a positive effect on the probability of price change. Durations have a negative effect. In-sample and out-of-sample forecasting analyses reveal that, in 71% of cases, the system successfully predicts the direction of the subsequent price change.

How do Individual UK Producer Prices Behave?

This article presents a number of stylised facts about the behaviour of individual UK producer prices. First, on average 26% of prices change each month, although there is considerable heterogeneity between sectors. Second, the probability of price changes is not constant over time. Third, the distribution of price changes is wide, although a significant number of changes are relatively small. Fourth, prices that change more frequently do so by less. Conventional pricing theories struggle to match these results, particularly the marked heterogeneity.

How Do Individual UK Producer Prices Behave?

This article presents a number of stylised facts about the behaviour of individual UK producer prices. First, on average 26% of prices change each month, although there is considerable heterogeneity between sectors. Second, the probability of price changes is not constant over time. Third, the distribution of price changes is wide, although a significant number of changes are relatively small. Fourth, prices that change more frequently do so by less. Conventional pricing theories struggle to match these results, particularly the marked heterogeneity.

Retail Price Setting in Uruguay

In recent years there has been a large increase in the empirical literature on price behavior. As new and detailed data sets become available, we observe a number of important studies on the microeconomic fundamentals of price setting by firms-mainly retailers-and their impact on inflation. This analysis allows a better understanding of the behavior, dispersion, and volatility of prices.In this paper, we use a rich and unique data set of 30 million daily prices in grocery stores and supermarkets across Uruguay to analyze stylized facts about consumer price behavior. Our findings are as follows:-The median duration of prices is two and one-half months. Therefore, retail prices in Uruguay are less sticky than in the United States and Brazil but stickier than in Chile and the United Kingdom.-We do not find evidence of a seasonal pattern in the likelihood of price adjustments.-The frequency of price adjustment is correlated with expected inflation only for the personal care product category. For the food category we find that supermarkets change the percentage points of the adjustment but not their frequency.-The probability of price change on the first day of the month is nine times higher than on any other day.-The probability of a price change is not constant over time.-There exists a high synchronization of price changes in our database, either at the city level or chain level. Overall, our analysis seems to be consistent with time-dependent models, although the high synchronization of price changes on the first day of the month awaits a better theoretical explanation.A Brief Review of the Empirical LiteratureAlthough there are different theoretical models in the literature that explain the microeconomic behavior of prices-such as menu cost models, sticky price and sticky information models, and time- or state-dependent pricing strategies-the stylized facts still avoid a unique theoretical explanation. Klenow and Malin (2010), which provides an up-to-date and concise overview of the empirical evidence, confronts the data with different theoretical models. The authors stress ten facts of the microeconomic behavior of prices. The primary facts are that prices do change at least once a year; that the main instrument for downward price adjustment is sales; that most markets have a stickier reference price; that goods prices differ in frequency of adjustment and the changes are asynchronous for different types of goods; that microeconomic forces explain the behavior of prices that differ from aggregate inflation; and that prices adjust mainly when wages change.Gopinath and Rigobon (2008) studies the stickiness of traded goods using microdata on U.S. import and export prices at the dock for the period 1994- 2005. The authors find long price duration for traded goods-10.6 months for imports and 12.8 months for exports; great heterogeneity in price stickiness across goods at the disaggregated level; a declining probability of price adjustment over time for imports; and a rather low exchange rate pass-through into U.S. import prices.Nakamura and Steinsson (2008) uses the consumer price index (CPI) and the producer price index (PPI) from the U.S. Bureau of Labor Statistics (BLS) for the period 1988-2005 to study price stickiness. The results show that there is a duration of regular prices of between eight and eleven months, after excluding sale prices; that temporary sales are an important source of price flexibility-mainly downward price flexibility; that roughly one-third of price changes, excluding sales, are price decreases; that price increases strongly covary with inflation, but price decreases do not; and that price changes are highly seasonal, mainly in the first quarter. Finally, the study finds that the hazard function of price changes, which estimates the probability of a price change after t periods without changing, is slightly downward sloping, which implies that the probability of a price change occurring decreases the longer the time span since the last change. …

How Do Individual UK Consumer Prices Behave?

This paper examines the behaviour of individual consumer prices in the United Kingdom, and uncovers a number of stylised facts about pricing behaviour. First, on average 19% of prices change each month, although this falls to 15% if sales are excluded. Second, the probability of price changes is not constant over time. Third, goods prices change more frequently than services prices. Fourth, the distribution of price changes is wide, although a significant number of changes are relatively small and close to zero. Fifth, prices that change more frequently tend to do so by less. We find that conventional pricing theories struggle to match these results, particularly the marked heterogeneity, which argues against the use of ‘representative agent’ models.

Do Supermarket Prices Change from Week to Week?

This paper examines the behaviour of supermarket prices in the United Kingdom, using weekly scanner data supplied by Nielsen. A number of stylised facts about pricing behaviour are uncovered. First, prices change very frequently in supermarkets, with 40% of prices changing each week, and even controlling for ‘temporary’ changes, a quarter of prices change each week. Importantly, there is evidence that focusing on monthly observations, rather than weekly ones, overstates the implied stickiness of prices. Second, the probability of price changes is not constant over time - all product categories have declining hazard functions. Third, the range of price changes is very wide, with some very large price cuts and price rises; but despite this, a significant number of price changes are very small. Fourth, there appears to be little link between the frequency and magnitude of price changes - prices that change less frequently do not tend to change by more. Fifth, the strongest correlation between price and volume changes is contemporaneous, suggesting that prices and volumes move together from week to week. And sixth, rough analysis based on simplifying assumptions suggests that consumers are fairly price sensitive: volumes change by more than prices.

Pricing Models: A Bayesian DSGE Approach for the U.S. Economy

This paper compares and estimates three pricing mechanisms in the context of a small DSGE model of the U.S. economy. We interpret our results as favoring the pricing mechanism presented in Wolman (1999 Wolman model) over the New Keynesian model with indexation (Gali and Gertler 1999, Smets and Wouters 2004a) and the sticky information model of Mankiw and Reis (2002). The key factor that explains the performance of the Wolman model is that the data reject the key assumption of the New Keynesian model that the firm's probability of price change is constant over time and independent of the contract's vintage. Our results also show that incorporating indexation in the New Keynesian model represents a poor expedient in matching the autocorrelation function of the inflation process over the last 20 years.

Predicting Stock Price Movements: An Ordered Probit Analysis on the ASX

Using stocks from a wide range of industry sectors on the Australian Securities Exchange, this paper examines the conditional distribution of intra-day stock prices and predicts the direction of the next price change in an ordered-probit-GARCH framework that accounts for the discreteness of prices. The analysis also incorporates the endogeneity of the time between trades in an ACD framework. Other elements considered include depth, trade imbalance, and volume. The results show that trade imbalance has a positive effect on the probability of price change. Durations have a negative effect. In-sample and out-of-sample forecasting analyses reveal that in 71% of the cases the system successfully predicts the direction of the subsequent price change.

Price Changes in the Euro Area and the United States: Some Facts from Individual Consumer Price Data

Prices of goods and services do not adjust immediately in response to changing demand and supply conditions. This paper characterizes the average frequency and size of price changes in the euro area and its member countries, investigates the determinants of the probability of price changes, and compares the evidence for the euro area with available U.S. results. The facts documented in this paper are based on evidence from individual price data recorded at the store level in all euro area countries except Ireland and Greece: that is in datasets covering Austria, Belgium, Finland, France, Germany, Italy, Luxembourg, the Netherlands, Portugal, and Spain, which together account for around 97 percent of euro area GDP. The data used are the monthly price records underlying the computation of national Consumer Price Indices and Harmonized Consumer Price Indices. These data cover a large number of products selected on the basis of extensive Household Budget Surveys.