betaFeedback

WorksMathematics

Scalable Amortized Variational Inference for Non-Poisson Buy-'Til-You-Die Models

Abstract

Despite the wide variety of existing Buy-`Til-You-Die (BTYD) models, nearly all rely upon the convenient assumption of transactions following a Poisson process. As modern customer bases grow larger and more diverse, a major gap in the marketing literature is BTYD models that can account for heterogeneity in timing patterns across millions of customers. This paper addresses that gap, introducing a family of models that assume transactions follow a Weibull renewal process and developing a highly scalable scheme for parameter estimation based on an amortized variational inference procedure. The proposed model fits to a proprietary dataset of 5 million online retail customers in 8 minutes which would take the current state-of-the-art an estimated 3-4 days. We show both theoretically and empirically that this dramatic improvement in computational performance comes with no appreciable change to either model interpretation or predictive performance. Beyond scalability, gradient-based variational inference also makes it easy to extend the model to covariates, which we illustrate on a public dataset of 4 million political donors during the 2020 US General election cycle. More generally, this paper demonstrates how to blend recent advances in approximate Bayesian inference and the tools of modern machine learning to dramatically improve the efficiency and expressivity of probabilistic models for customer base analysis.

Introduction

Large-scale customer relationship management (CRM) data recording the behavior of millions or more customers is now commonplace in industries like banking [48], telecommunications [80], and many others. Although increasingly large in length, CRM data often remains highly sparse at the level of individual customers, due in part to the common pattern of attrition by which customers only actively transact for short periods of time, after which they churn or “die” [63]. Large customer bases are moreover naturally heterogeneous, with many customers exhibiting a variety of different purchasing patterns [2]. This combination of size, sparsity, and heterogeneity motivates statistical models of CRM data that strike a particular balance of being 1) scalable, capable of being efficiently fit to millions of customers, 2) parsimonious, allowing for statistical information to be pooled across sparsely observed units, but also 3) sufficiently expressive, not collapsing salient and predictive heterogeneity in customer behavior.

Probabilistic models following the “Buy ‘Til You Die” (BTYD) framework of [68] are a mainstay of statistical marketing [37, 27], owing in part to their parsimonious representation of customer behavior. This representation splits cleanly into: 1) a transaction process, governing the timing by which a customer “buys”, and 2) a lifetime process, governing the time until a customer “dies”. A small set of such models—e.g., the original Pareto/NBD of [68], or the BG/NBD model of [30]—are simple enough to be fit scalably to large datasets using gradient-based optimization schemes which exploit such models’ analytic tractability. While these “workhorse” models account for a large share of practical usage, a large literature has since introduced many new variants which tend to trade analytic and computational tractability for expressivity, introducing more flexible probabilistic assumptions that then necessitate approximate inference schemes like MCMC for parameter estimation [58, 1, 59, 22].

Despite the wide variety of BTYD models to date however, very few depart from the analytically convenient assumption that the event times of an active customer’s transactions follow a Poisson process. The Poisson has long been the canonical model for “pure chance” events in many fields, with departures from it indicating the presence of unaccounted for structure. In marketing, transaction patterns that are either more regular or more “clumpy” than expected under a Poisson process have been highlighted and studied for several decades [15, 36, 92, 60]. Nevertheless, the lack of scalable non-Poisson models for heterogeneous patterns of timing in large-scale customer bases remains a major gap.

This paper fills that gap by introducing a family of non-Poisson BTYD models which enjoy almost all the core benefits of the “workhorse” models—namely 1) they can be scalably fit to millions of customers, 2) their key quantities can be efficiently computed, and 3) they can be easily extended to covariates. In particular, this paper shows how to extend amortized variational inference [45, 65, 89] to such models, which turns the problem of posterior inference into a gradient-based optimization problem, toward which all the tools of modern machine learning—i.e., GPU accelerated computation, automatic differentiation, and neural networks—can be deployed.

This family of models is built around transactions following a Weibull renewal process, an assumption with precedent in the marketing literature [15, 36, 3, 54] but one which departs from existing regularity-aware BTYD models which instead assume gamma-distributed renewals [60, 64]. As we show, the shift from gamma to Weibull facilitates the development of gradient-based variational inference, and does so without any appreciable change to such models’ expressivity, interpretation, or performance. We parameterize the Weibull in a nonstandard way so that the model emits a set of customer-specific latent variables with the same interpretation as in previous work, a correspondence which we characterize formally. In addition to the model, we also rely upon the Weibull distribution as the variational family during inference, where its convenient analytic properties contribute to a low-variance stochastic optimization procedure via pathwise or “reparameterization” gradient estimates [39, 45]. Due to this dual reliance on the Weibull distribution, we dub the proposed approach “Wei’ ‘Til You Die” (WTYD).

We illustrate the importance of modeling heterogeneous regularity in large-scale customer bases with two case studies on both proprietary and public data. The first case study is on 5 million customers of an online subscription-based retail firm, whose anonymized data we acquired from the marketing analytics firm Ocurate. The second is on 4 million individual political donors during the run-up to the 2020 US General elections, the data for which is publicly available through the Federal Elections Commission (FEC). In both cases, WTYD reveals interpretable heterogeneous structure in regularity, frequency, and churn that informs and accords with our knowledge of both domains. Fitting WTYD takes 8 minutes on the dataset of 5 million, which takes the current state-of-the-art an estimated 3-4 days. This comes with no appreciable loss in predictive performance, as we demonstrate in empirical comparisons on random subsets of the real data as well as on a suite of simulated datasets. On simulated data, where the ground-truth is known, we also show that WTYD exhibits no degradation in parameter or latent-variable recovery despite its dramatic improvement in computational efficiency. We also provide an illustration on how to extend WTYD to covariates and demonstrate the benefit in doing so.

Roadmap. The contributions of this paper are structured as follows.

  • After defining notation and preliminaries on CRM data in § 2, we give an overview of related BTYD models in § 3 which highlights a specific set of tradeoffs and connections.

  • We then introduce the WTYD model in § 4 where we formally characterize its close correspondence to the Pareto/GGG model of [59] and give several key properties, including its likelihood and a principled approximation to expected lifetime value.

  • In § 5, we then develop an amortized variational inference procedure that fits WTYD by stochastic gradient-based optimization. This section develops the basic procedure based on parameterizing an inference network tailored to point process observations as well as on pathwise gradient estimators. It also introduces a series of algorithmic improvements which are instrumental to making the basic algorithm work in practice, including Rao-Blackwellization, which exploits analytic properties of the model to decrease gradient variance, and “inference compilation” [46] which pretrains the inference network on synthetic data. Finally, this section introduces principled variational approximations to key posterior quantities such as “P(alive)”, the probability a customer is alive given their transactions.

  • In § 6, we report an extensive suite of simulation studies which compares WTYD to Pareto/GGG on recovering ground-truth parameters and latent-variables and forecasting future data across a range of data settings, showing that WTYD obtains a substantial speedup without any appreciable loss in performance.

  • In § 7 we then provide two case studies on a novel proprietary dataset of 5 million online retail customers and a public dataset of 4 million political donors. In both, we show that WTYD recovers interpretable heterogeneous structure in regularity, frequency and churn, doing so highly efficiently. We also provide an example of how to extend WTYD to covariates.

  • Finally, we conclude in § 8 by drawing further connections to the literature and outlining promising future avenues that build upon this work.

Customer Relationship Management (CRM) Data

CRM data consists of “which customer bought what and when” micro-records. As the timing pattern of such purchases is often the main object of statistical analysis, we will focus on the scenario where there is only one type of purchase to simplify notation and exposition.

A CRM database grows over time as new customers are added (or “born”) and their transactions recorded. At any current time TT, there are nn customers in the database, each of whom is indexed by i∈[n]i \in[n] where [n]≔{1,…,n}[n] \coloneqq\{1,\ldots,n\}. Let ti,0<Tt_{i,0}<T be the time that customer i∈[n]i\in[n] is “born”, often defined as the time of their first transaction. The times of their subsequent or “repeat” transactions are then ti≔{ti,1,…,ti,yi}\mathbf{t}_i \coloneqq\{t_{i,1},\ldots,t_{i,y_i}\} where ti,j<Tt_{i,j}<T for all j∈[yi]j\in[y_i]. The time of their latest transaction ti,yit_{i,y_i} is called their “recency” while the total count of their repeat transactions during the observation window yi∈Ny_i\in\mathbb{N} is called their “frequency”. Note that if yi=0y_i=0 then ti=∅\mathbf{t}_i=\emptyset.

For the purpose of evaluation and benchmarking of forecasting methods, we can also define a future holdout period up until time T⋆>TT^\star>T, such that the observation (or training) window is (0,T)(0,T) and the holdout (or testing) period is (T,T⋆)(T,T^\star). We then denote the transaction times of customer ii during the holdout period to be ti⋆≔{ti,1⋆,…,ti,yi⋆⋆}\mathbf{t}_i^\star\coloneqq\{t_{i,1}^\star,\ldots,t_{i,y_i^\star}^\star\} such that ti,j⋆∈(T,T⋆)t_{i,j}^\star\in(T,T^\star) for all j∈[yi⋆]j\in[y_i^\star] and where yi⋆y_i^\star denotes their number of transactions during the holdout period, which we call the customer’s “future frequency”. Note again that if yi⋆=0y_i^\star=0 then ti⋆=∅\mathbf{t}_i^\star=\emptyset.

Since customers are “born” into the data at different times, it is common to represent the timing of their transactions relativistically, which simplifies model specification. First, rather than deal in absolute timestamps, we consider the customer’s inter-transaction times (ITTs), which for yi>0y_i > 0 in the observation window are denoted Δti≔(Δti,1,…,Δti,yi)\Delta t_i \coloneqq(\Delta t_{i,1}, \ldots, \Delta t_{i,y_i}) and defined as

Δti,j≔ti,j−ti,j−1for j∈[yi] such that ti,j=ti,0+∑j′≤jΔti,j′.(1)\Delta t_{i,j} \coloneqq t_{i,j} - t_{i,j-1} \quad\text{for } j \in[y_i] \text{ such that } t_{i,j} = t_{i,0} + \sum_{j' \le j} \Delta t_{i,j'}. \tag*{(1)}

Similarly, the customer’s ITTs during the holdout period are then Δti⋆≔(Δti,1⋆,…,Δti,yi⋆⋆)\Delta t_i^\star\coloneqq(\Delta t_{i,1}^\star, \ldots, \Delta t_{i,y_i^\star}^\star) for yi⋆>0y_i^\star> 0, where the first of these is the gap between the customer’s last transaction during the observation window and their first transaction during holdout—i.e., Δti,1⋆≔ti,1⋆−ti,yi\Delta t_{i,1}^\star\coloneqq t_{i,1}^\star- t_{i,y_i}.

Considering only the ITTs is called the “customer’s journey” representation, which allows us to specify models that assume customers are “born” simultaneously at t=0t = 0. The length of each customer’s observation window then varies, each of which is defined Ti≔T−ti,0T_i \coloneqq T - t_{i,0} and often called their “total time”. By extension, Ti⋆≔T⋆−ti,0T_i^\star\coloneqq T^\star- t_{i,0} is their total time through the holdout period. For a visual representation of these quantities, see Figure 1.

Figure 1. Raw CRM data (left) based on absolute transaction timestamps versus the “customer journey” representation (right) based on inter-transaction times and varying total times.

Buy `Til You Die (BTYD) Models

A common pattern of consumer behavior is churn where customers eventually cease purchasing from a firm permanently. Accounting for this behavior is important—e.g., for valuing a customer base or even an entire firm [37, 53]—especially in non-contractual settings where the firm never directly observes a cancellation event [63]. Marketers have traditionally done so by modeling each customer as being active (or “alive”) during a set interval (ti,0,τi)(t_{i,0}, \tau_i)—i.e., between the time of their “birth” ti,0t_{i,0} and “death” τi\tau_i. Critically, while a customer’s birth is observed in CRM data, their death typically is not. Death manifests indirectly though, such as a long-churned customer having “recency” much less than the current time ti,yi≪Tt_{i,y_i} \ll T. Thus, statistical models that treat τi\tau_i explicitly as a latent variable are highly motivated.

A long history of such models in marketing, which collectively form the subfield of customer-base analysis [27], follows the “buy ‘til you die” framework of [68], wherein each model is fundamentally characterized by two processes:

  1. a transaction process for the timing of transactions ti\mathbf{t}_i while customers are “alive”, and

  1. a lifetime (or dropout) process for the time τi\tau_i until customers “die”.

Transaction renewal processes. The vast majority of BTYD models assume transactions follow a Poisson process, or more generally a renewal process [33, 17], under which ITTs are drawn Δti,j∼iidfi\Delta t_{i,j} \overset{\mathrm{iid}}{\sim} f_i from some renewal distribution fif_i specific to customer ii. For instance, a Poisson transaction process is one where renewals are Exponential—i.e., fi≔Exp⁡(RATE⁡=λi)f_i \coloneqq\operatorname{Exp}(\operatorname{RATE} = \lambda_i). This then induces a continuous-time renewal counting process Yi(⋅)Y_i(\cdot):

Δti,j∼iidfiandYi(t)≔max⁡{m:∑j=1mΔti,j≤t}.(2)\Delta t_{i,j} \overset{\mathrm{iid}}{\sim} f_i \quad\text{and} \quad Y_i(t) \coloneqq\max\left\{m : \sum_{j=1}^{m} \Delta t_{i,j} \le t\right\}. \tag*{(2)}

The customer’s “frequency” (as defined in the previous section) is then yi=Yi(Ti∧τi)y_i = Y_i(T_i \wedge\tau_i) where a∧b≔min⁡{a,b}a \wedge b \coloneqq\min\{a,b\}. The driving assumption behind BTYD models—i.e., that a customer never buys again after τi\tau_i—is then expressed as the “customer’s lifetime value” (CLV)1 being Yi(∞)=Yi(τi)Y_i(\infty)=Y_i(\tau_i). We can also define the counting process on any interval (s,t)(s,t) for s<ts<t to denote Yi(s,t)≔Yi(t)−Yi(s)Y_i(s,t) \coloneqq Y_i(t)-Y_i(s). For instance, a customer’s “future frequency” is yi⋆=Yi(Ti,Ti⋆)y_i^\star=Y_i(T_i,T_i^\star).

Key quantities. Practitioners have traditionally sought to query BTYD models in a few key ways, which can be cleanly divided into prior vs posterior quantities, the former being:

(1a) Expected customer lifetime value: E[Yi(∞)]\mathbb{E}[Y_i(\infty)],

(1b) Marginal likelihood: p(Di)=∫p(Di,zi) dzip(\mathcal{D}_i)=\int p(\mathcal{D}_i,z_i)\,\mathrm{d}z_i,

where Di≔{Δti,Ti,…}\mathcal{D}_i\coloneqq\{\Delta t_i,T_i,\ldots\} are the data on customer ii during the observation window, usually their ITTs and total time, but potentially other data (e.g., absolute timestamps tit_i, covariates xi\mathbf{x}_i), and zi≔{fi,τi,…}z_i\coloneqq\{f_i,\tau_i,\ldots\} are customer-specific latent variables, usually just their renewal distribution (or parameters thereof) and time of death. The key posterior quantities are then:

(2a) Forecasted future frequency: E[Yi(Ti,Ti⋆)∣Di]\mathbb{E}[Y_i(T_i,T_i^\star)\mid\mathcal{D}_i],

(2b) Posterior predictive distribution: p(Di⋆∣Di)=∫p(Di⋆,zi∣Di) dzip(\mathcal{D}_i^\star\mid\mathcal{D}_i)=\int p(\mathcal{D}_i^\star,z_i\mid\mathcal{D}_i)\,\mathrm{d}z_i,

(2c) Probability that customer ii is alive “P(ALIVE)”: p(τi>Ti∣Di)p(\tau_i>T_i\mid\mathcal{D}_i),

where Di⋆≔{Δti⋆,Ti⋆,…}\mathcal{D}_i^\star\coloneqq\{\Delta t_i^\star,T_i^\star,\ldots\} are the data on customer ii during the holdout period.

Being able to efficiently compute these quantities facilitates the use of such models. The expected frequencies in (1a) and (2a) are interpretable summary statistics used by marketers to segment the customer base and make predictions, with expected CLV in (1a) being particularly useful when customer-level covariates are available to help overcome the “cold-start” prediction problem. Tractable computation of the prior and posterior predictive distributions in (1b) and (2b) then facilitates parameter estimation and model selection, as it does with most probabilistic models. Finally, P(ALIVE) in (2c) plays a particularly central role in BTYD models, as computing it is a subroutine for computing most other key quantities, and it is itself an interpretable and useful summary statistic useful.

Whether the quantities above are tractable depends on the probabilistic assumptions of a given model. In the rest of this section, we survey some well-known BTYD models, emphasizing those we build upon, and highlighting the tension between tractability and expressivity that has been a key theme in this literature and which this paper seeks to ameliorate.

Pareto/NBD. The original BTYD model of [68] assumed that transactions follow a Poisson process until a Pareto-distributed time-of-death—i.e.,

τi∼Pareto⁡(shape=s(τ),scale=r(τ))(3)\tau_i \sim\operatorname{Pareto}(\mathrm{shape}=s^{(\tau)},\mathrm{scale}=r^{(\tau)}) \tag*{(3)}
Δti,j∼iidGamma⁡(shape=1,rate=λi)(4)\Delta t_{i,j} \mathrel{\overset{\mathrm{iid}}{\sim}} \operatorname{Gamma}(\mathrm{shape}=1,\mathrm{rate}=\lambda_i) \tag*{(4)}

References

  1. [1]M. Abe. “Counting your customers” one by one: A hierarchical Bayes extension to the Pareto/NBD model. Marketing Science, 28(3):541–553, 2009.
  2. [2]G. M. Allenby and P. E. Rossi. Marketing models of consumer heterogeneity. Journal of Econometrics, 89(1–2):57–78, 1998.
  3. [3]G. M. Allenby, R. P. Leone, and L. Jen. A dynamic model of purchase timing with application to direct marketing. Journal of the American Statistical Association, 94(446):365–374, 1999.
  4. [4]L. Ambrogioni, U. Güçlü, J. Berezutskaya, E. van den Borne, Y. Güçlütürk, M. Hinne, E. Maris, and M. van Gerven. Forward amortized inference for likelihood-free variational marginalization. In AISTATS, pages 777–786, 2019.
  5. [5]J. Ansel et al. PyTorch 2: Faster Machine Learning Through Dynamic Python Bytecode Transformation and Graph Compilation. In ASPLOS, 2024.DOI
  6. [6]E. Ascarza, O. Netzer, and B. G. S. Hardie. Some customers would rather leave without saying goodbye. Marketing Science, 37(1):54–77, 2018.
  7. [7]L. J. Bain and M. Engelhardt. Probability of correct selection of Weibull versus gamma based on livelihood ratio. Communications in statistics-theory and methods, 9(4):375–381, 1980.DOI
  8. [8]E. P. Batislam, M. Denizel, and A. Filiztekin. Empirical validation and comparison of models for customer base analysis. International Journal of Research in Marketing, 24(3):201–209, 2007.
  9. [9]C. Bauckhage. Computing the Kullback-Leibler divergence between two generalized gamma distributions. arXiv preprint arXiv:1401.6853, 2014.arxiv.org/abs/1401.6853
  10. [10]D. M. Blei, A. Kucukelbir, and J. D. McAuliffe. Variational inference: A review for statisticians. Journal of the American Statistical Association, 112(518):859–877, 2017.DOI
  11. [11]J. Bornschein and Y. Bengio. Reweighted wake-sleep. In ICLR, 2015.
  12. [12]L. Bottou, F. E. Curtis, and J. Nocedal. Optimization methods for large-scale machine learning. SIAM Review, 60(2):223–311, 2018. doi: 10.1137/16M1080173.DOI
  13. [13]L. Bouton, J. Cagé, E. Dewitte, and V. Pons. Small campaign donors. Working Paper 30050, National Bureau of Economic Research, 2022.
  14. [14]H. Castéran, L. Meyer-Waarden, and W. Reinartz. Modeling customer lifetime value, retention, and churn. In Handbook of Market Research, pages 1001–1033. Springer, 2021.
  15. [15]C. Chatfield and G. J. Goodhardt. A consumer purchasing model with Erlang inter-purchase times. Journal of the American Statistical Association, 68(344):828–835, 1973.DOI
  16. [16]W. Chen, B. Chen, Y. Liu, Q. Zhao, and M. Zhou. Switching Poisson gamma dynamical systems. In IJCAI, 2020.
  17. [17]D. R. Cox. Renewal Theory. Methuen, 1962.
  18. [18]C. Cremer, X. Li, and D. Duvenaud. Inference suboptimality in variational autoencoders. In ICML, pages 1078–1086, 2018.
  19. [19]C. Davidson-Pilon. lifetimes: Predictive Customer-Lifetime Value in Python, 2020. URL https://pypi.org/project/lifetimes/. Python package, MIT License.
  20. [20]W. Dean, J. Orduz, C. Allen, and C. E. Trujillo. PyMC-Marketing: Bayesian marketing toolbox in PyMC, 2026. URL https://www.pymc-marketing.io. Python package.
  21. [21]A. P. Dempster, N. M. Laird, and D. B. Rubin. Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society: Series B (Methodological), 39(1):1–22, 1977.
  22. [22]R. Dew and A. Ansari. Bayesian nonparametric customer base analysis with model-based visualizations. Marketing Science, 37(2):216–235, 2018.
  23. [23]Z. Duan et al. Sawtooth factorial topic embeddings guided gamma belief network. In ICML, pages 2903–2913, 2021.
  24. [24]L. Dziurzynski, E. Wadsworth, and D. McCarthy. BYTD: Implementing buy ’til you die models. http://CRAN.R-project.org/package=BYTD, 2014.
  25. [25]P. S. Fader and B. G. Hardie. A note on deriving the Pareto/NBD model and related expressions, 2005.
  26. [26]P. S. Fader and B. G. Hardie. Incorporating time-invariant covariates into the Pareto/NBD and BG/NBD models, 2007.
  27. [27]P. S. Fader and B. G. Hardie. Probability models for customer-base analysis. Journal of Interactive Marketing, 23(1):61–69, 2009.
  28. [28]P. S. Fader and B. G. Hardie. The Gamma-Gamma model of monetary value. https://www.brucehardie.com/notes/025/gamma__gamma.pdf, February 2013. Technical note (unpublished).
  29. [29]P. S. Fader, B. G. Hardie, and K. L. Lee. “Counting your customers” the easy way: An alternative to the Pareto/NBD model. Marketing Science, 24(2):275–284, 2005a.
  30. [30]P. S. Fader, B. G. Hardie, and K. L. Lee. Implementing the BG/NBD model for customer base analysis in Excel, 2005b.
  31. [31]P. S. Fader, B. G. S. Hardie, and J. Shang. Customer-base analysis in a discrete-time noncontractual setting. Marketing Science, 29(6):1086–1108, 2010.
  32. [32]D. Fearn and E. Nebenzahl. On the maximum likelihood ratio method of deciding between the Weibull and Gamma distributions. Communications in Statistics-Theory and Methods, 20(2): 579–593, 1991.
  33. [33]W. Feller. An Introduction to Probability Theory and Its Applications, Vol. II. Wiley, 1971.
  34. [34]M. Figurnov, S. Mohamed, and A. Mnih. Implicit reparameterization gradients. Advances in Neural Information Processing Systems, 31, 2018.arxiv.org/abs/1805.08498
  35. [35]S. Gershman and N. Goodman. Amortized inference in probabilistic reasoning. In CogSci, 2014.
  36. [36]S. Gupta. Stochastic models of interpurchase time with time-dependent covariates. Journal of Marketing Research, 28(1):1–15, 1991.
  37. [37]S. Gupta et al. Modeling customer lifetime value. Journal of Service Research, 9(2):139–155, 2006.
  38. [38]P. Hall, T. Pham, M. P. Wand, and S. S. Wang. Asymptotic normality and valid inference for Gaussian variational approximation. The Annals of Statistics, 39(5):2502–2532, 2011.DOI
  39. [39]S. G. Henderson and B. L. Nelson. Handbooks in Operations Research and Management Science: Simulation, volume 13. Elsevier, 2006.
  40. [40]J. Herniter. A probabilistic market model of purchase timing and brand selection. Management Science, 18(4-part-ii):P102–P113, 1971.DOI
  41. [41]G. E. Hinton, P. Dayan, B. J. Frey, and R. M. Neal. The “wake-sleep” algorithm for unsupervised neural networks. Science, 268(5214):1158–1161, 1995.
  42. [43]K. Jerath, P. S. Fader, and B. G. Hardie. New perspectives on customer “death” using a generalization of the Pareto/NBD model. Marketing Science, 30(5):866–880, 2011.
  43. [44]J. L. Katzman, U. Shaham, A. Cloninger, J. Bates, T. Jiang, and Y. Kluger. DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network. BMC Medical Research Methodology, 18:24, 2018.
  44. [45]D. P. Kingma and M. Welling. Auto-Encoding Variational Bayes. In ICLR, 2014.
  45. [46]T. A. Le, A. G. Baydin, and F. Wood. Inference compilation and universal probabilistic programming. In AISTATS, pages 1338–1348, 2017.
  46. [47]T. A. Le, A. R. Kosiorek, N. Siddharth, Y. W. Teh, and F. Wood. Revisiting reweighted wake-sleep for models with stochastic control flow. In UAI, pages 1039–1049, 2020.
  47. [48]R. A. d. L. Lemos, T. C. Silva, and B. M. Tabak. Propension to customer churn in a financial institution: A machine learning approach. Neural Computing and Applications, 34(14):11751–11768, 2022.
  48. [49]G. Lorden. On excess over the boundary. The Annals of Mathematical Statistics, 41(2):520–527, 1970.
  49. [50]I. Loshchilov and F. Hutter. Decoupled weight decay regularization. In ICLR, 2019.
  50. [51]D. B. Magleby, J. Goodliffe, and J. A. Olsen. Who Donates in Campaigns? The Importance of Message, Messenger, Medium, and Structure. Cambridge University Press, 2018.
  51. [52]C. C. Margossian and D. M. Blei. Amortized variational inference: when and why? In UAI, pages 2434–2449, 2024.
  52. [53]D. M. McCarthy and P. S. Fader. Customer-based corporate valuation for publicly traded noncontractual firms. Journal of Marketing Research, 55(5):617–635, 2018.
  53. [54]B. McShane, M. Adrian, E. T. Bradlow, and P. S. Fader. Count models based on Weibull interarrival times. Journal of Business & Economic Statistics, 26(3):369–378, 2008.DOI
  54. [55]H. Mei and J. M. Eisner. The neural Hawkes process: A neurally self-modulating multivariate point process. In NeurIPS, 2017.
  55. [56]J. Näf, R. V. Mbelson, and M. Meierer. CLVAE: A Variational Autoencoder for Long-Term Customer Revenue Forecasting. arXiv preprint arXiv:2604.22636, 2026.arxiv.org/abs/2604.22636
  56. [57]R. M. Neal and G. E. Hinton. A view of the EM algorithm that justifies incremental, sparse, and other variants. In Learning in graphical models, pages 355–368. Springer, 1998.DOI
  57. [58]O. Netzer, J. M. Lattin, and V. Srinivasan. A hidden Markov model of customer relationship dynamics. Marketing Science, 27(2):185–204, 2008.
  58. [59]M. Platzer. Customer base analysis with BTYDplus, 2016.
  59. [60]M. Platzer and T. Reutterer. Ticking away the moments: Timing regularity helps to better predict customer activity. Marketing Science, 35(5):779–799, 2016.DOI
  60. [61]R. Pyke. Markov renewal processes: definitions and preliminary properties. The Annals of Mathematical Statistics, 32(4):1231–1242, 1961.DOI
  61. [62]R. Ranganath, S. Gerrish, and D. Blei. Black box variational inference. In AISTATS, pages 814–822, 2014.arxiv.org/abs/1401.0118
  62. [63]W. J. Reinartz and V. Kumar. On the profitability of long-life customers in a noncontractual setting: An empirical investigation and implications for marketing. Journal of Marketing, 64(4):17–35, 2000.DOI
  63. [64]T. Reutterer, M. Platzer, and N. Schröder. Leveraging purchase regularity for predicting customer behavior the easy way. International Journal of Research in Marketing, 38(1):194–215, 2021.DOI
  64. [65]D. J. Rezende, S. Mohamed, and D. Wierstra. Stochastic backpropagation and approximate inference in deep generative models. In ICML, pages 1278–1286, 2014.arxiv.org/abs/1401.4082
  65. [66]H. Robbins and S. Monro. A stochastic approximation method. The Annals of Mathematical Statistics, 22(3):pp. 400–407, 1951.
  66. [67]D. C. Schmittlein and D. G. Morrison. Prediction of future random events with the condensed negative binomial distribution. Journal of the American Statistical Association, 78(382):449–456, 1983.
  67. [68]D. C. Schmittlein, D. G. Morrison, and R. Colombo. Counting your customers: Who are they and what will they do next? Management Science, 33(1):1–24, 1987.DOI
  68. [69]D. A. Schweidel and G. Knox. Incorporating direct marketing activity into latent attrition models. Marketing Science, 32(3):471–487, 2013.
  69. [70]D. A. Schweidel and W. W. Moe. Binge watching and advertising. Journal of Marketing, 80(5):1–19, 2016.DOI
  70. [71]O. Schur, A. C. Türkmen, T. Januschowski, and S. Günnemann. Neural temporal point processes: A review. arXiv preprint arXiv:2104.03528, 2021.
  71. [72]S. Squires, A. Prügel-Bennett, and M. Niranjan. A variational autoencoder for probabilistic non-negative matrix factorisation. arXiv preprint arXiv:1906.05912, 2019.arxiv.org/abs/1906.05912
  72. [73]D. Sridhar, H. Daumé III, and D. Blei. Heterogeneous supervised topic models. Transactions of the Association for Computational Linguistics, 10:732–745, 2022.
  73. [74]A. Srivastava and C. Sutton. Autoencoding variational inference for topic models. In ICLR, 2017.
  74. [75]M. Titsias and M. Lázaro-Gredilla. Doubly stochastic variational Bayes for non-conjugate inference. In ICML, pages 1971–1979, 2014.
  75. [76]D. Titterington and B. Wang. Convergence properties of a general algorithm for calculating variational Bayesian estimates for a normal mixture model. Bayesian Analysis, 1(3):625–650, 2006.
  76. [77]V. A. Traag. Complex contagion of campaign donations. PLOS ONE, 11(4):e0153539, 2016.DOI
  77. [78]U.S. Federal Election Commission. Bulk campaign finance data: Contributions by individuals, 2019–2020 cycle. https://www.fec.gov/data/browse-data/, 2021. File indiv20, accessed June 2026.
  78. [79]J. Valendin, T. Reutterer, M. Platzer, and K. Kalcher. Customer base analysis with recurrent neural networks. International Journal of Research in Marketing, 39(4):988–1018, 2022.
  79. [80]W. Verbeke, D. Martens, and B. Baesens. Social network analysis for customer churn prediction. Applied Soft Computing, 14:431–446, 2014.
  80. [81]M. J. Wainwright, M. I. Jordan, et al. Graphical models, exponential families, and variational inference. Foundations and Trends® in Machine Learning, 1(1–2):1–305, 2008.
  81. [82]C. Wang, X. Liu, D. Wang, H. Zhang, B. Chen, and M. Zhou. Cross-scale Weibull graph attention autoencoder for modeling document networks. arXiv preprint arXiv:2410.09696, 2024.arxiv.org/abs/2410.09696
  82. [83]C. Wang et al. Multimodal Weibull variational autoencoder for jointly modeling image-text data. IEEE Transactions on Cybernetics, 52(10):11156–11171, 2022.DOI
  83. [84]W. Weibull. A statistical distribution function of wide applicability. Journal of Applied Mechanics, 18(3):293–297, 1951.
  84. [85]J. S. White. Weibull renewal analysis. Society of Automotive Engineers (SAE), 1964.DOI
  85. [86]S. Wiegrebe, P. Kopper, R. Sonabend, B. Bischl, and A. Bender. Deep learning for survival analysis: a review. Artificial Intelligence Review, 57:65, 2024.
  86. [87]C. J. Wu. On the convergence properties of the EM algorithm. The Annals of Statistics, 11(1):95–103, 1983.
  87. [88]N. Yannaros. Weibull renewal processes. Annals of the Institute of Statistical Mathematics, Vol. 46, No. 4:641–648, 1994.
  88. [89]C. Zhang, J. Bütepage, H. Kjellström, and S. Mandt. Advances in variational inference. IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(8):2008–2026, 2019.
  89. [90]H. Zhang, B. Chen, D. Guo, and M. Zhou. WHAI: Weibull Hybrid Autoencoding Inference for Deep Topic Modeling. In ICLR, 2018.
  90. [91]S. Zhang, X. Fan, B. Chen, and M. Zhou. Bayesian attention belief networks. In ICML, pages 12413–12426, 2021.
  91. [92]Y. Zhang, E. T. Bradlow, and D. S. Small. Predicting customer value using clumpiness: From RFM to RFMC. Marketing Science, 34(2):195–208, 2015.DOI
  92. [93]Z. Zhou, X. Yang, R. Rossi, H. Zhao, and R. Yu. Neural point process for learning spatiotemporal event dynamics. In L4DC, pages 777–789, 2022.
  93. [94]S. Zuo, H. Jiang, Z. Li, T. Zhao, and H. Zha. Transformer Hawkes process. In ICML, pages 11692–11702, 2020.

Paper details

vlnc:W_ju38mng5
Version
v1 · latest
Published
Added
License
CC BY 4.0
Usage

views · 2

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Mathematics

Not recorded for this paper

View statement
Looking for related papers…
Contents