Could network analysis of horizontal visibility graphs be faithfully used to infer long-term memory properties in real-world time series?

Contemporary literature suggests that the effective market hypothesis is not substantiated. Instead, it suggests the Fractal Market Hypothesis (FMH). Fractal markets are characterized by long-term memory. The main feature of the fractal market is that the frequency distribution of the indicator looks the same across diffe rent investment horizons. In such cases, it is said that for an appropriate indicator, the phenomenon of scale invariance is observed. All daily changes are correlated with all future daily changes, all weekly changes are correlated with all future weekly changes. There is no characteristic time scale, a key characteristic of the time series. The presence of memory in the time series can be characterized by the Hearst indicator. This paper analyzes the hryvnia to US dollar exchange rate for the period 04.06.14-04.01.15. Finding the Hearst index made it possible to conclude that there is or is not long-term memory in this series. The presence of long-term memory indi cates that the efficient market hypothesis is unjustified. The hypothesis was tested that the longer the averaging intervals are taken into account in the model, the Hearst's index decreases. The analysis does not have great predictive power, however, it allows to identify the presence or absence of long-term memory in the study process and thus to accept or reject the hypothesis of an effective market. That is, the series under study is persistent, thus demonstrating long-term me mory availability. Thus, since persistence is revealed, the hypothesis of an effective market for the exchange rate yield is not confirmed, but instead can be argued for the fractality of the hryvnia / dollar exchange rate yield. Therefore, the application of the proposed approach made it possible to find the Hearst rate for the hryvnia / dollar exchange rate. The value found indicates that the effective market hypothesis is not substantiated for at least such an exchange rate.

Download Full-text

Comparing Visual Memories to Novel Visual Input Risks Lasting Memory Distortion

10.31234/osf.io/xr4su ◽

2021 ◽

Author(s):

Joseph M. Saito ◽

Katherine Duncan ◽

Keisuke Fukuda

Keyword(s):

Working Memory ◽

Young Adults ◽

Real World ◽

Eyewitness Testimony ◽

Long Term Memory ◽

Memory Distortion ◽

Term Memory ◽

Memory Accuracy ◽

Memory Distortions

Maintaining perceptual experiences in visual working memory (VWM) allows us to flexibly accomplish various tasks, but some tasks come at a price. For example, comparing VWM representations to novel perceptual inputs can induce inadvertent memory distortions. If these distortions persist, they may explain why everyday memories often become unreliable after people perform perceptual comparisons (e.g., eyewitness testimony). Here, we conducted two experiments to assess the consequences of perceptual comparisons using real-world objects that were temporarily maintained in VWM (n = 32) or recalled from long-term memory back into VWM (n = 30). In each experiment, young adults reported systematic memory distortions following perceptual comparisons. These distortions increased in magnitude with the delay between encoding and comparisons and were preserved when memories were retrieved again a day later. These findings suggest that perceptual comparisons play a mechanistic role in everyday memory distortions, including situations where memory accuracy is vital.

Download Full-text

Working memory is not fixed capacity: More active storage capacity for real-world objects than simple stimuli

10.31234/osf.io/dk89w ◽

2020 ◽

Author(s):

Timothy F. Brady ◽

Viola S. Störmer ◽

George Alvarez

Keyword(s):

Working Memory ◽

Real World ◽

Visual Information ◽

Memory Task ◽

Memory Systems ◽

Long Term Memory ◽

Term Memory ◽

Active Storage ◽

Working Memory Models

Visual working memory is the cognitive system that holds visual information active to make it resistant to interference from new perceptual input. Information about simple stimuli – colors, orientations – is encoded into working memory rapidly: in under 100ms, working memory ‘fills up’, revealing a stark capacity limit. However, for real-world objects, the same behavioral limits do not hold: with increasing encoding time, people store more real-world objects and do so with more detail. This boost in performance for real-world objects is generally assumed to reflect the use of a separate episodic long-term memory system, rather than working memory. Here we show that this behavioral increase in capacity with real-world objects is not solely due to the use of separate episodic long-term memory systems. In particular, we show that this increase is a result of active storage in working memory, as shown by directly measuring neural activity during the delay period of a working memory task using EEG. These data challenge fixed capacity working memory models, and demonstrate that working memory and its capacity limitations are dependent upon our existing knowledge.

Download Full-text

ERROR ASSESSMENT IN MODELING WITH FRACTAL BROWNIAN MOTIONS

Fractals ◽

10.1142/s0218348x13500187 ◽

2013 ◽

Vol 21 (03n04) ◽

pp. 1350018 ◽

Cited By ~ 4

Author(s):

BINGQIANG QIAO ◽

SIMING LIU

Keyword(s):

Time Series ◽

Fractal Dimension ◽

Hurst Exponent ◽

Long Term Memory ◽

Error Assessment ◽

Data Sampling ◽

Brownian Motions ◽

Term Memory ◽

Rescaled Range

To model a given time series F(t) with fractal Brownian motions (fBms), it is necessary to have appropriate error assessment for related quantities. Usually the fractal dimension D is derived from the Hurst exponent H via the relation D = 2-H, and the Hurst exponent can be evaluated by analyzing the dependence of the rescaled range 〈|F(t + τ) - F(t)|〉 on the time span τ. For fBms, the error of the rescaled range not only depends on data sampling but also varies with H due to the presence of long term memory. This error for a given time series then can not be assessed without knowing the fractal dimension. We carry out extensive numerical simulations to explore the error of rescaled range of fBms and find that for 0 < H < 0.5, |F(t + τ) - F(t)| can be treated as independent for time spans without overlap; for 0.5 < H < 1, the long term memory makes |F(t + τ) - F(t)| correlated and an approximate method is given to evaluate the error of 〈|F(t + τ) - F(t)|〉. The error and fractal dimension can then be determined self-consistently in the modeling of a time series with fBms.

Download Full-text